773 9.6-4) dovia7 7 y ~0.148y° = 0.773 —0.148(0.773°) = 0.705 For steam flow in Ib/hr, 83C,Cr (y - 0.148)" (9.6-6) (1+ 0.00077 ,) 16,130 = 1.83C(0.8)(34.7)(0.705) = Cy = 450 gpm/psi"? To design for 100% overcapacity, we need a maximum Cyn = 2Cy = 950 gpm/psi', From Figure 9.6-3, a 10-in. Masoneilan valve, with a coefficient of 1000, is the smallest valve that can provide the service. Using Fisher Controls equation, we have fn = 1.060, faR wel 88 fe] GVA | Saturated steam at 20 psig, Tsui ~ 259°F, 2; = 0.0833 Ib/f?. Let Cy = 35, we have 59.64 3 16,130 = 1.06, (0. ys a Cy = 15,000 i OIF oo 3s V3aF Ee . For 100% overcapacity Cy max = 2(15,000) = 30,000 . The corresponding Cy is then = CglCy = 30,000/35 = 856 gpmipsi!” ‘We obtain similar value for the valve coefficient C, from both methods. 9-26Chapter 10 Flow in Chemical Engineering Application 10.1 Drag Force on Solid Particles in Fluids Vi ‘Turbulent eddies at high Nr. Figure 10.1-1 Flow past a sphere. When a fluid flows past a spherical particle it will exert a net total drag force Fp on the sphere in the direction of the flow as shown in Figure 10.1-1, This total drag force consists of the viscous drag and the form drag. The viscous or frictional drag is the results of the ‘momentum transfer between the fluid and the sphere due to the difference in their velocities. ‘The form drag is the net force exerted on the sphere due to the difference in pressure upstream and downstream of the particle. The total drag force J" can be determined from the following equation _ FIA, Cp (10.1-1) Hi ateta J pW. gfe This equation is the definition of the drag coefficient Cp where 4y is the projected area of the particle in the direction of motion, For a sphere 4) = nD*/4, The definition of the drag coefficient from equation (10.1-1) is analogous to that of the friction factor for flow in a pipe T ~. (10.1-1) rae Jn this equation, tw is the force exerted on the wall by the fluid per unit area. For laminar flow (Nae = PDV0/y1 < 1) the differential momentum equation can be solved for the flow over the sphere to obtain the pressure and local stress distributions. The integration of the stress distribution over the sphere surface gives the viscous drag and the integration of the pressure over the sphere surface gives the form drag. For this case the drag coefficient is given as 10-124. Co = 5 For Ngo > 1, the drag coefficient can be obtained from experiment. A simple correlation, which represents the entire range of Cy vs. Ne reasonably well up to about Npe = 2x10°, is given by Dallavalle' for fhuids flow over a sphere: Cp = (00s ae } (10.1-3) For flow past a circular cylinder with L/D >>1 normal to the cylinder axis, the flow is similar to that over a sphere, The drag coefficient over the entire range of Nr (up to 210°) is given by 10.2 Separations by Free Settling ‘Many engineering processes require the separation of solid particles from fluids. Separation might be achieved by free settling under gravity or other forces since particles of different size and density will move through the fluid at a different rate, As a first step, we will determine the velocity of the particle in free settling or the travel time of a particle through a given distance. Consider a spherical particle with diameter D and density p, shown in Figure 10.2-1, which is settling under gravity in a fluid of density p and viscosity x. Figure 10.2-1 Settling of a spherical particle under gravity. Application of a downward momentum balance on the particle gives d (aD'p, £ Ly | =Fy-Fp-F 10.21 4 (Ber -m FoR 102-1) where ‘Darby, R., Chemical Engineering Fluid Mechanics, Marcel Dekker, 2001, p. 341 10-2Fg= gravity force = Dog Fp = drag force = ; Cy p¥"Ay, 2 Fy = buoyant force = wee If there is no change in particle velocity, equation (10.2-1) becomes (102-2) ‘The velocity obtained from solving equation (10.2-2) is called the terminal velocity V, when the gravity force exactly balances the buoyant and drag forces and there is no acceleration. Example 10.2-1? - — A stainless steel sphere of diameter D = 1 mm and density p, = 7,870 kg/m’ falls steadily under gravity through a polymeric fluid of density p = 1,052 kg/m? and viscosity = 0.1 kg/m’s. Determine the downwards terminal velocity of the sphere. Solution ——— At terminal velocity, there is no acceleration of the particle Solving for the velocity yields pu Dale, <2 a 4 Delp, - Pp) (1) aoe Cp ‘We need an expression for the drag coefficient Cp to solve for the velocity, we could use equation (10.1-3) Cp = (oe (10.1-3) However, using this expression will result in a nonlinear equation. A simpler method that might work is to assume laminar flow so that * Wilkes, James, Fluid Mechanics for Chemical Engineers, Prentice Hall, 1999, pg. 215, 10-3Cp = 24 2 24H "Ne PVD Substituting this expression into equation (E-1) gives re D'g(p,- p) _ (10%)* (9.81(7,870—1,052) 0.0372 ms 184 18)(0.1) We now need to check the assumption of laminar flow: Nae = DV! 44 = (1,052)(0.0372)(0.001/0.1 = 0.39 <1 ‘The flow is laminar hence the terminal velocity is 3.72 em/s. Example 10.2-2° A spherical hot air balloon of diameter 40 ft and deflated mass 500 Ib is released from rest in still air at 50°F and 1 atm. The gas inside the balloon is effectively air at 200°F and 1 atm. Assuming a constant drag coefficient of Cp = 0.60, estimate (a) The steady upward terminal velocity of the balloon, (b) The time in seconds it takes to attain 99% of this velocity. Solution ~ MP RT Air density can be estimated from ideal gas law: p: 29)(14.7) A At50°F and 1 atm, pi = 2204-1) — 9, anc PL 0.73510) 0,078 Ib/ft _29)(14.7) = 0.060 Ib/ft? (10.73)(660) At 200°F and 1 atm, 2 = Application of the momentum balance on the balloon gives av M= =F,-Mg-F; b— Mg— Fp In this equation Mis the total mass of the balloon. The buoyant force is evaluated as 2 2 PDD, g = EA WON) (59 2) = 94,165 Ibs? Re ECG 6 > Wilkes, James, Fluid Mechanies for Chemical Engineers, Prentice Hall, 1999, pg. 216 10-4‘The total mass of the balloon is 00 + BANOS) 2511 b The gravity force is Mg = (2,511)(32.2) = 80,842 Ib-ft/s* ‘The drag force is Fo= 5 Cp pV"Ay = 30 {6)(0.078)"(xt)(20") = 29.47" (a) The steady upward terminal velocity of the balloon, 29.402 = Fy — Mg = 84,165 — 80,842 = 3,323 Ib-fus? wn ne (2) = 10.6 fs 29.4 (b) The time in seconds it takes to attain 99% of this velocity. 2s = Fo- Mg Fy = 3,323 - 29.47 4 =13-7 dt essa” 113-77 0.997; = 10.5 Using the integration formula { 1 113"? +105 = 854— in US 2Vi13 113"? -10.5 =20.48 Solid particles can be removed from a dilute suspension by passing the suspension through a Jarge vessel as shown in Figure 10.2-2. In the vessel, if the upward velocity of the fluid (Q/4) is less than the terminal velocity (7) of the particles and the residence time is long enough, the particles will settle out, 10-5Figure 10.2-2 Gravity settling tank. For laminar flow, the terminal velocity of the particles is given by Betas y= Dee, -p) (10.1-4) 184 This velocity must be greater than the upward velocity of the liquid, Therefore D'g(p,- p) 184 ‘The diameter of the smallest particle that will settle out is 1a D= (243) (10.1-5) (slo, -p)4. If the flow is not laminar, we can divide the correlation for the drag coefficient (10.1-3), by Nee to obtain 1 ‘This equation can be rearranged to a quadratic equation in term of —— Nae 48 | 0.632 -(2) a Ne Nis Wee 10-6Solving for Te yields = (10080 +0, 208) Vee Once the Reynolds number is determined, the diameter of the smallest particle that will settle out can be evaluated 12 } = 0.0658 ke Nye = VPP. => p = Hee = HN ev PQ We first need to determine the ratio C2, From the force balance on the spherical particle at ne terminal velocity 0=Fy~Fo-Fi aD'p, 1 xD'p "2, 1 6 pa,- 2, G 8~ 7 Copy ~ Fg aD*(p, - p)g _ D*(p,-p)g _ AD(e, - p)g C= Pn 3pV7A, 3pV aD? 14 3pV* We can divide the above equation by Reynolds number to eliminate the unknown D Lp - Aue, ~ eg (10.1-6) Nie 3p? ‘The procedure to determine the smallest particle that will settle out is as follows. 1) Calculate the terminal velocity 7 = 2 Lup, 2) Caleulate ARLE 3p°v? 3) Calculate {000s + 0.208 = 0.0658 Vr ) 4) Caloulate D = #Nae pv 10-7If the problem is to determine the maximum flow rate Q so that particles of a given size D will settle out we can follows the following procedure: 1) Solve for Cp from the force balance on the particle 3pV7A, 3pV? aD? /4 3pVv" 2) The unknown in with Nas" is case is the terminal velocity that can be eliminated by multiplying Cp ee CoNg2 = 220-2) (=) 4D*plp.- pig _ 4 Ne 3pV? Hw 3u° (oss 3) This is a quadratic equation in term of Nge!, therefore Nee" = (14.42-+1.827YN,) - 3.798 4) Caleulate the terminal velocity from Reynolds number Nag = VPP oy y= Hee # 5) The maximum volumetric flow rate so that particles of a given size D will settle out is then =v ~ Mee =vA= Ne 4 Q “D 10-8Chapter 10 10.3 Flow Through Packed Beds Air flow Figure 10.3-1 Flow through a packed bed, A packed bed is simply a column or tube packed with solid particles as shown in Figure 10.3- 1. Flow through packed beds occurs in many engineering application for example the fluid flow through a tubular reactor containing catalyst particles or the water flow through cylinders packed with ion-exchange resin in order to produce deionized water. Flow through a packed bed is a sub area of the flow through a porous media in which a solid, or collection of solid particles, has sufficient open space in or around the solids to enable a fluid to pass through or around them. Since the fluid in a packed bed follows a tortuous path through the interstices or pores between particles, one method of modeling the flow behavior is to consider the flow path as a noncircular conduit with a hydraulic diameter defined as A Volume _of voids Dy=4 Sa a eae Wetted surface area a D4 ex Bed volume aoa) "(No.of particles\Surface areal Particle) ‘ 10-9‘The bed porosity, & which is the fraction of total volume that is void is defined as volume voids volume of entire bed volume of entire bed -volume_of _ particles volume of entire bed (10.3-2) Idealized pore Particles Fluid path Figure 10.3-2 A model of the flow through pores. Figure 10.3-2 shows a model for the flow through the pores of the particles in a packed bed. All the empty volume within the bed is considered as the volume 4jE of a noncircular conduit with surface area Wp representing the surface of contact between the fluid and the solid particles. The number of particles N, in the bed may be estimated from the average particle volume N, (Bed volume)(Fraction of _ solids bed) (Bed _volume)( aoe Volume! Particle Substituting equation (10.3-3) into equation (10.3-1) yields Dynan (2) (103-4) particle surface area In this equation a, = . If the particles are spherical with diameter D, particle volume then ge ee “DD 6 10-10For a bed composed of uniform spherical particles with diameter D _ _2De : ues) ce When the packing has a shape different from spherical, an effective particle diameter is defined particle volume ___ 6 _ yp. (103-6) Ona or aaais aac rey. particle surface area a, In this equation Ds is the diameter of a sphere with the same volume as the particle and ¥'is the sphericity or shape factor defined by Surface area of a_sphere ith same volume as the particle Pp, P, Figure 10.3-3 A horizontal packed bed with idealized pores, Applying the energy equation over a horizontal packed bed shown in Figure 10.3-3 yields (10.3-7) (10.3-8) where roe is the friction factor in the pores and is the interstitial or actual velocity within the pores. The actual velocity relates to the approach or superficial velocity V, by the expression lO =2.% . ae (103-9) Substituting the hydraulic diameter, equation (10.3-5), and the interstitial velocity, equation (10.3-9), into equation (10.3-8) yields oe eae v,y ,30-s)_,, ,20-«) L = YroroViP — = Yoore| — ] LA" = Shore’ a rarest pa iaie (®) 2D,e FD, 10-11From equation (10.3-7), the pressure drop across the bed is given by al — AP =~ (Po — Pi) = per = Yporep¥s Rearranging the equation so that the friction factor in the pore is on the right-hand side, we have ‘The numerical factor 3 is normally absorbed in the fiction factor fyar. There for the above equation is usually written as pv? L ime For turbulent flow, the friction factor free was found experimentally to be a constant so that the above equation becomes ap D, pv? L ime = 175 For laminar flow, the friction factor fyac, also found experimentally, is expressed in terms of Reynolds number by an equation similar to that in laminar flow in a pipe 180 Sree Naegore ‘ PY,D, f The pore Reynolds number is defined as Nrajox = ¢7—E An expression that adequately ae represents the pressure drop in both laminar and turbulent flow is - = 1) sas Nre,pore ‘This equation with a value of 150 instead of 180 is called the Ergun equation. 10-12Example 103-1 AA packed bed is composed of cylinders having a diameter D = 002 mand a length h =D. The bulk density of the overall packed bed is 962 kg/m’ and the density of the solid cylinder is 1600 kg/m’, Calculate the void fraction ¢ and the effective diameter Dp. Solution ~ Let the volume of the packed bed = 1 m°, The total mass of the bed is then (962 kg/m®)(1 m°) or 962 kg. This is also the mass of the solid cylinders. The volume of the cylinders or particles contained in the bed is Volag = SOE <4, = 0.601 m* 1600kg / m ‘The void fraction # is then volume of entire _bed-volume_of _ particles volume of entire bed Zp particle volume _ Pitt Foe wie oF yaad Example 10.3.2 —— Air at 311°K is flowing through a packed bed of spheres having a diameter of 12.7 mm, The void fraction ¢ of the bed is 0.38. Find the pressure drop across the bed. Bed diameter = 0.61 m P, = 110aim Mass flow rate = 0.358 kg/s H=244m Solution ~ ‘We can use the Ergun equation to determine the pressure drop through the packed bed D, VD, Ae eaiie 150175 Npepor ® Lt PVE L 1-6” Nog oy 10-13150V,wL(l-e)’ 1.75pV,L0.~6) Dpe® Dye" —AP= In this equation, the density should be evaluated at the 0.5(P1 + P2). However since P; is not known, we will evaluate the density at the inlet pressure pee RT 1G) 1.248 ke/my Note: R= 8314 m*-Pa/kmol-°K_ The superficial velocity is y= = th. (0358) ___ 9.981 mis pA pad? (1.248)(7)(0.61)? For air at 311°K and 1.1 atm: = 1.9x10° kg/ms ie 150H0=8) «1 a5py, } (0.981)2.44)(1=0.38) [(150)0.9x10°)0=038) «751 248)(0.981) (0.0127)(0.38)° 0.0127 — AP = 4.86x10° Pa = AP =~ (P2 — P}) = 4.8610" Pa P2 =P, ~ 4.86x10° = 1.10x1.013«10° ~ 4,86x10° = 1.114%10° Pa Since P, is close to P; (less than 1% difference), we do not need to reevaluate the air density. The pressure drop across the bed is ~ AP =4,86x10° Pa 10-14Chapter 10 10.4 Laminar Flow Through Porous Medium ‘The Ergun equation can also be applied to consolidated medium such as a brick or a sandstone rock formation through which oil is flowing. For laminar flow at AP Py te 150 15001 —e) yr WEL I-8 Nu PVD, Solving for the superficial velocity yields __ AP Die? Ly 1500-7 B Pp = 10.4-1 T 0.4-1) Ys Equation (10.4-1) is called d’Arcy’s law with the permeability «defined by Pea 1500 ‘The most common unit for the permeability is the darcy, which is defined as the flow rate in ‘cm*/s that results when a pressure drop of 1 atm is applied to a porous medium that is 1 cm* in cross-sectional area and 1 cm long, for a fluid with viscosity of 1 cP. For this case the superficial velocity is 1 cm/s, therefore Com XeP) — 9.98610 om? 1 darcy = aim/cm The differential form of d'Arcy’s law in one-dimensional flow is x dP = KS 10.4.2) ude ¢ ) Ve : Kk In this expression, v, is a flux (a volumetric flow rate per unit area), “is a conductance, and H ¢ is a driving force so that the equation has the general form: Fux = Driving force. _ Conductance x Driving force Resistance For three-dimension flow, equation (10.4-2) becomes v= ~ VP (10.4-3) “ ‘Substituting this equation into the continuity equation for incompressible flow yields 10-15Vv=V:-=vP=0 # For constant “, the equation is simplified to H VVP=V'P=0 (10.4-4) Numerical Solution of the Laplace equation using COMSOL Multiphysics Flow in two-dimensional porous medium is governed by Laplace equation in terms of pressure as: @P oP oF Once the pressure is known, the velocity vector can be evaluated from K K OP aP = Evpory=- £2, y= 5 © where xis the permeability # Be” ne @P Pg an = p=t0 p=0 > © ap Py an Solve the problem’ of flow in @ porous medium shown in the above figure. The flow region is a rectangle ABCD with a central circular region E of zero permeability that allows no flow through i. The boundaries AB and CD ar also impervious to flow so that 2 = 0, The n * Wilkes, James, Fluid Mechanics for Chemical Engincers, Prentice-Hall, 1999, p.566 10-16pressure along the two ends are P = 10 psig and P = 0 psig. Let “ = 1.0 tb-fi/s throughout H the region. A partial differential equation solver such as COMSOL Multiphysics developed by COMSOL Inc could solve the set of differential and algebraic equations encountered in fluid mechanics. The ability of this software can be found from the COMSOL User’s Guide: “COMSOL Multiphysics is designed to make it as easy as possible to model and simulate physical phenomena, With COMSOL Multiphysies, you can perform free-form entry of custom partial differential equations (PDEs) or use specialized physics application modes. ‘These physics modes consist of predefined templates and user interfaces that are already set up with equations and variables for specific area of physics. Further, by combining any number of these application modes into a single problem description, you can model a ‘multiphysies problem-such as one involving simultaneous mass and momentum transfer.” COMSOL Multiphysics can solved a variety of problems beside fluid mechanics, including those in heat transfer, chemical reactor design, electromagnetics, fuel cells, transport phenomena, polymer processing and so on. COMSOL is very flexible with a friendly interface. The Graphical User Interface consists of the following steps: 1. Model Navigator: Use this dialog box to choose the model equation, its dimensions, stationary or time-dependent analysis. 2. Geometry Modeling: Use this interface to draw or specify object and its dimensions and coordinates. 3. Physics Settings: The boundary and initial conditions are specified in the Boundary Settings. The physical properties in the model equations are specified in the Subdomain Settings. Mesh Generation and Solution: The mesh can be initialized and refined if necessary. The problem is then solved with various options for solver like: stationary linear, stationary nonlinear, time dependent and so on. The solver can also be auto selected. 5. Postprocessing and Visualization: The results can be displayed in different graphical or numerical formats. Getting Started 1) Start the program by going through the Start Menu and double-clicking on Comsol Multiphysics v3.x. Doing so opens the Model Navigator Dialog box. This is the starting point for the definition of the certain types of physics phenomena of interest. Note: The software provides default values for each step therefore you might not have to choose a value for some of the steps explained in the following intructions. 2) Click on the New Tab in the Model Navigator. 3) In the Space Dimension, select (2D) to simulate the two-dimensional flow. 4) Under the Application Modes folder, single-click on the + sign for PDE Modes folder. Single-click + sign for the Classical PDEs, Single-click the + sign again for Laplace’s Equation. 10-175) The Model Navigator shows the dependent variable, which is to be solved for. Enter Pas the dependent variable. 6) Click OK. REPMe as [rv |e Pe] pS fe Hosein nin fe yentenunans ee Se Syren toe fe Syanctodechnes |B Strocmae 2 Syanloes © canetencoentasion Storrs 8 ese oo ee Ca Geometry Modeling 7) The next step in the modeling process is to create the model geometry. In this case, simulate the reactor as a rectangle. To do this, select Specify Object on Draw menu. Select Rectangle. 8) On the dialog box, enter 2.5 for width, and 1.8 for height. Choose Center for the Base Postion. Click OK. 9) Select Specify Object on Draw menu. Select Circle. On the dialog box, enter 0.5 for radius. Click OK, 10-1810) Since we want the domain to be a rectangle with a circular hole in it, Select Create Composite Object on Draw menu. Edit the Set formula field to contain: RI - C1 10-19 Zoom Extents12) The following sereen will appear. Boundary Conditions 13)Select Boundary Settings on Physics menu. Choose Dirichlet boundary condition for Boundary 1 and a value of 10 for r (since P = 10). Choose Dirichlet boundary condition for Boundary 4 and a value of 0 for r (since P = 0). Choose Newmann boundary condition for Boundaries 2, 3, 5, 6, 7, and 8 since 2° = 9 for these boundaries. 10-2014) Select Plot Parameters on Postprocessing menu. On the General tab, check Surface, Contour, Arrow, and Geometry edges. Click on the Arrow tab and edit “— Px” for x component and“ Py” for y component since velocity is defined by tie | 15)Choose Initialize Mesh from the Mesh menu. Next choose Refine Mesh to improve the first rough mesh. This option can be repeated to reduce the mesh size further. Everything is now ready for generating a solution. Choose Solve Problem from the Solve ‘menu. The results will be generated automatically and the values of the dependent variable will be color-coded in the solution domain, A color bar at the right assigns numerical solution values to each color. The isobar and velocity vector are also plotted. 10-21To summarize, Comsol Multiphysics allows you to use drawing tools to create solution domains. You can then choose the PDE to be solved, assign PDE parameters appropriate for the domain, assign boundary conditions to boundary segments, and specify initial conditions for the PDE if necessary. You can then generate triangular meshes of different refinements, compute discrete solutions at the nodes of the mesh, and display high-quality plots of the continuous approximation to the PDE solution over the domain and even over times. 10-22Chapter 10 10.5 Flow Through Fluidized Beds ‘The upward flow of fluid through a bed of solid particles is an important process occurring in nature and in industrial operations. At low velocities, the pressure drop in a horizontal bed increases with the fluid velocity according to the Ergun equation Lae 175 (10.5-1) pore AP = pressure drop through the packed bed ee) particle diameter p= fluid density V, = superficial velocity at a density averaged between inlet and outlet conditions 6 — =bed porosity pV D, (=e) Nie,pore = = average Reynolds number based upon actual velocity ‘The overall energy balance over the horizontal bed is BP ter=0 p ‘Therefore the frictional dissipation term per unit mass flowing is AP _ 150uVh (ls), 1.758V7h 10.5-2) ppp; dD, ee When the packing has a shape different from spherical, an effective particle diameter is defined as particle volume _ 6 particle surface area a, = ps (10.5-3) In this equation Ds is the diameter of a sphere with the same volume as the particle and V is the sphericity or shape factor defined by Surface area of a sphere with same volume _as_the particle ‘Surface area of the particle 10-23Equation (10.5-3) can also be written as N, (particle volume) total particle volume Eye cae eee ROAR) nite pariicte vole tee PY''N, (article. surface area) total particle surface area In this equation, N, is the number of particles in the bed. (total particle _volume)(1—«) (total particle surface area —e) Dp (bed _volumeX1—e) (total particle surface area) Dp (10.5-4) In this equation A, is the interfacial area of packing per unit of packing volume, fi7/ft’ or m’/m’. The bed porosity is defined as volume of entire bed volume of _ particles volume of entire bed aren —ei8h of _all_particles ‘particle density aR*h (105-5) where R = inside radius of column, , and ¢ are characteristics of the packing. Experimental values of can easily be determined from Eq, (10.5-5) but A, for non-spherical particles is usually more difficult to obtain, Values of 4, and sare available for the common commercial packing in various references. 4, for spheres can be computed from the volume and surface area of a sphere. ‘As the gas velocity inoreases, conditions finally ocour where the force of the pressure drop times the cross-sectional area just equals the weight of the particles in the bed. A slight increase in gas velocity, to increase the pressure drop, is required to unlock the intermeshed fixed-bed particles. Once the particles disengage from each other, they begin to move. The pressure drops to the point where the upward force on the bed is balanced by the downward force due to the weight of the bed particles. Further inoreases in gas velocity fluidize the bed, the pressure drop rises slightly until slugging and entrainment occurs. The point of maximum pressure drop shown in Figure 10.5-1 is the point of minimum fluidization. Applying the energy equation for a vertical bed at this point yields Pi Po= pest patna = hall Gn2)Po + &tP)] & (10.5-6) where har= bed height at the minimum fluidization point. 10-24Equation (10.5-6) is rearranged to Perm a(t ~ G0lpy ~ 2) (105-7) Ig Pressure point of fluidization drop AP, Inches of water Log Re Figure 10.5-1 Pressure drop in a bed of particulate solids. Eq. (10.5-2) can be rearranged with D, substituted by yD,, V, substituted by Vax and substituted by Aneto obtain 150 ay by TS AVR (en) cos3 gaye Dares WD, by The minimum fluid velocity Vig at which fluidization begins can be calculated by combining Eqs. (10.5-7) and (10.5-8) to obtain the quadratic equation in terms of Ving 1530p (lee, 150M ay (Im zap) su = g(l- a)(Py- P) (105-98) ar = wlPp- Pp) (10,5-9b) For many systems, Equation (10.5-9b) simplifies to: Di(p, - D, Vat= Pile, ele for Neane = | Ph.) <20 (10.5-10a) 1650y 7 and 10-25x jg = Pee PM fo ays > 1000 205-106) 24.50 At high fluid velocities, when the expansion of the bed is large, the behavior of fluidization depends on whether the fluid is a liquid or a gas. With a liquid, fluidization is smooth and uniform without large bubbling. This kind of fluidization is known as “particulate” fluidization. With a gas, uniform fluidization is frequently observed only at low velocities. ‘At high velocities, non-uniform or “aggregative” fluidization will be observed with large bubbling and the bed is then often referred to as a “boiling” bed, In long, narrow fluidized beds, coalescence of the bubbles might be large enough to cover the entire cross-section of the column, These slugs of gas altemate with slugs of fluidized solids are carried upwards and subsequently collapse, causing the solids to fall back again, Slugging can cause severe entrainment problems and hence is undesirable. InaP Fluidized bed Inv, Figure 10.5-2 Variation of bed height with superficial velocity Figure 10.5-2 shows the relation between the actual bed height / and the superficial velocity V,. For low velocity, the bed height / remains constant and the bed functions like a packed bed. However, as V is increased, the pressure drop will eventually high enough to counter balance the downward weight of the particles. This is the point of minimum fluidization where the velocity has reached the incipient fluidizing velocity Vy. From this point on, the bed behave as a fluidized bed where the bed continues to expand with increasing velocity while the pressure drop increases only slightly. Example 10.5-1 Solid particles having a size of 0.12 mm, a sphericity shape factor y of 0.88, anda density of 1000 kg/m’ are to be fluidized using air at 2 atm and 25°C. The voidage (void fraction) at minimum fluidizing conditions is 0.42. (1) If the cross sectional area of the empty bed is 0.30 m* and the bed contains 300 kg of solid, calculate the minimum height fgg of the fluidized bed. (2) Determine the pressure drop at the minimum fluidizing conditions. (3) Determine the incipient fluidizing velocity Vne. 10-26Solution -— (1) Calculate the minimum height lige of the fluidized bed. ‘The mass of the solid particles in the bed is given by mg = PoAlgd | ~ Ent) => Fmt = 300 (1000)(0.3)(1- 0.42) Ire (2) Determine the pressure drop at the minimum fluidizing conditions. ‘The pressure drop is balanced by the gravity and buoyant forces on the particles (APA) = Ahins) [C. ~ End Pp + Se?) & Air at 2 atm and 25°C, p= 2.374 kg/m? AP = (9.81)(1.724)[(1 — 0.42)(1000) + (0.42)(2.374)] = 9,827 Pa (3) Determine the incipient fluidizing velocity Vii. We can solve equation (10.5-b) for the incipient fluidizing velocity Viyr L.75p¥z, , (l-e,,) 150uY, Pat Nf) 0M ar op, — p) (10.5-9) wD, a wD: 8(Po - P) ) For air at 2 atm and 25°C, = 1,85x10° Pas. We also have wD, = (0.88)(0.12x10") = 1056x104 m. Using numerical values in equation (10.5-9b), we have 2 5 (7512.37 + G20) 1500.85 x10" oy ~9.81(1000 -2.374) (1.056 x 10*)(0.42) 0.427 (1.05610) 5.31x10°(Vint)? + 1.9481 10°F ne — 9.786710" = Voor = 5.0210" m/s 10-27Example 10.5-2 -- A® horizontal water purification unit consists of a hollow cylinder that is packed with ion- exchange resin particles. Tests with water flowing through the unit gave the following results: Flow rate Q(gallons/t) 10.0200 Pressure drop,—AP (psi) 4.0 10.0 If the available pump limits the pressure drop over the unit to a maximum of 54 psi, what is the maximum flow rate of water that can be pumped through it? Solution ‘The pressure drop across the packed bed can be determined by the Ergun equation pV .D, (eye Pp 2 10 15 Naepere = For a given packed bed, the only two variables are the pressure drop (~ AP) and the flow rate (Q). The Ergun equation can be rearranged to — AP = a, + BV? = aQ + 6 In this equation, a and b are parameters that depend on the characteristics of the packed bed. ‘They can be obtained from the data of pressure drop versus flow rate: 4= 10a + 100 10 = 20a + 4006 Solving for @ and b we obtain a ~03 psi/(gallons/hr) and a = 0.01 psil(gallons/hr)? ‘Therefore the pressure drop for this particular bed can be correlated by AP =030 + 0.010" For a maximum pressure drop of 54 psi, we have 54 = 0.3Qaux + 0.01 Q max? => Onnx? + 30Qnax — 5,400 = 0 This is a quadratic equation in terms of Qnax, hence Qian = — 15 + (15? + 5,400)°* = 60 gpm. ilkes, James, Fluid Mechanics for Chemical Engineers, Prentice-Hall, 1999, p:218 10-28Appendix A Previous Exams (2007) CHE31101 MOMENTUM TRANSFER QUIZ #1 (50 minutes) Not units, ‘our answers must be correct to 3 significant figures and have the appropriate I. When a hydrometer floats in water, its cylindrical Fig. Hydrometer stem is submerged so that a certain point on the é stem is level with the free surface of water, as shown in Fig. 1. When the hydrometer is placed in . : another liquid L of specific gravity s, the stem rises : so that point. is now 0.6 cm above the free surface Water Liquid L L. Determine the specific gravity s. If needed, the cross-sectional area of the stem is 0.1 cm’, and when in water a total volume of 1 cm’ (stem plus bulb) is submerged. Specific gravity = hem IL. A, Absolute pressures and temperatures must be employed when using the ideal gas law. B. To convert from psia to psig, add 14.7, approximately. a.AandBaretrue b, Only Ais true ¢.OnlyBistruc — d. A and Bare false IIL. A spherical hot-air balloon of diameter 15 m and deflated mass 400 kg is released in still air at 10°C. The gas inside the balloon is effectively air at 130°C. The density of air (kg/m’) is 1.24 at 10°C and 0.872 at 130°C. Acceleration of gravity is 9.81 m/s”. The buoyancy force on the balloon is Deism pnt FyepVeg — Fy= 214910 ww