Ch 8: Steady – State Non-isothermal Reactor Design Energy Balances, Rationale and Overview Calculate the volume necessary to achieve a conversion, X, in a PFR for a firstorder, exothermic reaction carried out adiabatically. For an adiabatic, exothermic reaction the temperature profile might look something like this: The combined mole balance, rate law, and stoichiometry yield: To solve this equation we need to relate X and T. We will use the Energy Balance to relate X and T. For example, for an adiabatic reaction, energy balance can be written in the form set X, then calculate T, -VA, and , increment X, then plot , the vs. X 1 1. Adiabatic (Q = 0); CSTR, PFR, Batch and PBR: 2. CSTR with heat exchanger, UA(Ta-T) and large coolant flow rate. 3 . PFR/PBR with heat exchange 3A. In terms of conversion, X 3B. In terms of molar flow rates, Fi 2 4. For Multiple Reactions 5. Coolant Balance These are eq’ns we will use to solve rxn engineering problems with heat changes. 3. Energy Balance Typical units for each term are J/s; i.e. Watts 3 Energy Balance 1st Law of Thermodynamics: dE = dQ – dW Energy of the system (for a closed system) Heat flow to the system Work done by the system on the surroundings dE • • = Q− W + Fin E in - Fout E out 123 dt {for an open system} rate of energy added to the system by mass flow Evaluation of the Work Term: • W = Wf + Ws flow shaft n n ~ = −∑ Fi ⋅ P ⋅ V i i =1 dt • in i =1 + Ws out ~ Hi = Ui + P Vi Combining with dE sys ~ + ∑ Fi ⋅ P ⋅ V i • n n i =1 i =1 = Q+ W s − ∑ Fi ⋅ H i in + ∑ Fi ⋅ H i out 4 (1) 1. 2. 3. 4. 5. 6. Replace Ei by Ei=Hi-PVi Express Hi in terms of enthalpies of formation and heat capacities Express Fi in terms of either conversion or rates of reaction Define ∆HRX Define ∆CP Manipulate so that the overall energy balance is either in terms of the Equations above 1.A, 1.B, 2, 3A, 3B, or 4 depending on the application Step 1: Substitute , and into equation (1) to obtain the General Energy Balance Equation. General Energy Balance: For steady state operation: 5 We need to put the above equation into a form that we can easily use to relate X and T in order to size reactors. To achieve this goal, we write the molar flow rates in terms of conversion and the enthalpies as a function of temperature. We now will "dissect" both Fi and Hi. Flow Rates, Fi For the generalized reaction: (2) In general, Fi = FA0 (Θi +ν i X) ν A = -1 ν B = - b a where Θi = νC = c a νD = Fi0 FA0 (3) d a Then n ∑H i =1 n i0 ∆H RX ∑F i0 ⋅ Fi0 - ∑ H i ⋅ Fi = FA0 [(H A0 - H A ) + (H B0 - H B ) Θ B ] + ........... i =1 ⎤ n n ⎡ = FA0 ∑ [(H i0 - H i ) Θ i ] - ∑ ⎢(ν i ⋅ H i ) FA0 ⋅ X ⎥ 14 2 43 ⎥ i =1 i =1 ⎢ ⎦ ⎣ ∆H RX d c b = HD + HC - HB - HA a a a H i0 − ∑ Fi H i = FA0 ∑ Θi (H i0 - H i ) - ∆H RX (T) FA0 X (4) (5) (6) ⎧ dE ⎫ for st - st ⎨ = 0⎬; combining (1) and (6) : ⎩ dt ⎭ • • n Q - W + FA0 ∑ Θi (H i0 - H i ) - ∆H RX (T) FA0 X = 0 (7) i =1 If a phase change takes place, it will be included in ∆H RX . 6 Assuming no phase change: H i = H i (TR ) + ∆H Qi o Enthalpy of formation at TR (ref. temp.) (8) Change in enthalpy when T is changed from TR to T. If there is a phase change: TM T TR TR H i = H i (TR ) + ∫ C ps,i ⋅ dT + ∆H mi (Tm ) + ∫ C pl,i dT o T2 ∆H Qi = ∫ C pi ⋅ dT (if no phase change) (9) (10) T1 T H i = H i (TR ) + ∫ C pi ⋅ dT o (11) TR Cp = α i + βi ⋅ T + γ i ⋅ T 2 T H i − H i0 = ∫C pi ⋅ dT = C pi ⋅ [T - Ti0 ] (12) Ti0 Substitute into (7) : • • Q - W + FA0 n ∑Θ i =1 i ⋅ C pi ⋅ [T - Ti0 ] - ∆ H RX (T) FA0 X = 0 (13) Combine eq’n (11) & (5) ∆H RX = ∆H RX + ∆C p (T - TR ) o (14) d c b o o o o H D (TR ) + H C (TR ) - H B (TR ) - H A (TR ) a a a (15) where ∆ H RX = o ∆C p = d c b C pD + C pC - C pB - C pA a a a (16) 7 Combine (13) and (14) • • [ n ] Q - W + FA0 ∑ Θi ⋅ C pi ⋅ [T - Ti0 ] - ∆H RX + ∆C p (T - TR ) ⋅ FA0 X = 0 i =1 o (17) Adiabatic Operation: • ⎫ Ws = 0⎪ ⎬ • Q = 0 ⎪⎭ X= [ ∑Θ C i o pi (T - Tio ) - ∆HRX + ∆Cp (T - TR ) ] for an exothermic, adiabatic rxn X for an exo. rxn why does X increase? This is from E balance, not mole balance. ∆H RX > ∆C p (T - TR ) o T so X vs T is linear!!! Adiabatic Tubular Reactor Rearrange (18): T= [ ] X ⋅ - H RX + ∑ Θ i ⋅ C pi To + X ⋅ ∆C p ⋅ TR o ∑Θ ⋅C i pi + X ⋅ ∆C p (19) Combine with differential mole balance: FA0 dX = - rA (X, T) dV (20) To obtain T, X and conc’n profiles along the reactor! Use (19) to construct a table of T vs X. Obtain k(T) as a function of X Æ -rA as a function of X. Look at Table 8-2A. 8 Steady – State Tubular Reactor with Heat Exchange: • • ∆ Q = U ⋅ ∆A (Ta - T) = Ua ⋅ (Ta - T) ⋅ ∆V FA0 ∑F ⋅H i To V • ∆ Q+ ∑ Fi ⋅ H i V − ∑ Fi ⋅ H i ∑F ⋅H T i assume Ws = 0 i FAe i Te V+∆V V + ∆V =0 (1) • The heat flow to the reactor ∆ Q Overall heat transfer coefficient (U) Heat exchange area (∆A) Difference between ambient temperature (Ta) and reactor temperature (T) • ∆ Q = U ⋅ ∆A (Ta - T) = Ua ⋅ ∆V ⋅ (Ta - T) A 4 = V D (2) Diameter of reactor If ∆V Æ 0 d ∑ (Fi ⋅ H i ) Ua (Ta - T) = (3) dV mole balance : dFi = ri = ν i (-rA ) dV (4) diff. Eq' n (3) Ua (Ta - T) = ∑ dFi dV H i = C pi ⋅ dT ⋅ H i - ∑ Fi dH i dV (5) dT dH i = C pi ⋅ dV dV (6) Combine (5) & (4) & (6) dT Ua (Ta - T) = ∑ν i ⋅ H i ⋅ (-rA ) - ∑ Fi ⋅ C pi 1 4 2 43 dV (7) ∆H RX 9 Æ Rearrange: 6Heat 4 generated 7 48 6 Heat 4 7removed 48 dT rA ⋅ ∆H RX − Ua (T - Ta) = dV ∑ Fi ⋅ Cpi (8) Fi = FA0 ⋅ (Θ + ν i X) Substitute into (8) dT Ua (Ta - T) + rA ⋅ ∆H RX = dV FA0 ⋅ (∑ Θ i + ν i X + ∆C p ⋅ X ) Ua for a PBR {dW = ρb dV} ρ dT = b dW {for a PFR} ⋅ (Ta - T) + (rA ') ⋅ ∆H RX ∑F ⋅C i These eq’ns will be coupled with mole balance eq’ns pi dX - rA = dV FA0 Balance on the Coolant Heat Transfer Fluid: Heat transfer fluid Tao R2 R1 FA0, To V V + ∆V Reactants The fluid will keep the rxn temperature constant for endo/exo – thermic rxns You might have A. Co – current Flow B. Counter – current Flow 10 A. Co – Current Flow The Energy Balance • • E in Ta: coolant temperature mc, Hc • + Q conduction = 0 - E out FA, T V + ∆V V • • + Ua (T - Ta) ∆V = 0 mc Hc - mc Hc FA, T V V + ∆V V mc, Hc V + ∆V Divide by ∆V and take limit as ∆V Æ 0 • mc ⋅ dH c + Ua (T - Ta) = 0 dV dH c dT = Cp a dV dV ⇒ dTa Ua (T - Ta) = • dV mc ⋅ Cp Tao exothermic Tao endothermic V B. Counter – Current Flow Ta2 Ta FAo, To T V dTa Ua (Ta - T) = • dV m c ⋅ C pc V + ∆V V=0 V = Vf At the entrance X = 0; V = 0; Ta = Ta2 At the exit V = Vf; Ta = Tao The sol’n to the counter-current flow problem to find {T X} is a trial & error procedure. Assume a coolant temp at the entrance (Ta2) Solve ODEs to calculate X, T and Ta as a function of V: Find Ta(V = Vo) If Ta (V = Vo ) - Ta2 < ε → Ta (V = Vo ) = Ta2 Else assume another Ta2 11 Equilibrium Conversion As T X {For endothermic rxns} As T X {For exothermic rxns} Exothermic Rxns: Xe = Kc 1 + Kc {1st order rxn} From Le Chaltlier’s Law (Kc as T Kc = if ∆H<0) exothermic Kp (RT )γ To find the max X in an exothermic rxn carried adiabatically: Energy balance To1 >To Xe Xe1 To Xe = ∑Θ i ⋅ C pi ⋅ (T - To ) − ∆H RXN (T ) ⎧if To is changed to To1 ⎫ ⎨ ⎬ Xe ⎩as T ⎭ To1 Adabatic temperature ~ ⎫ ⎧ dlnK o H (T) H ∆ ∆ RXN (TR ) + ∆ C p (T - TR ) p RXN ⎪ ⎪ = = ⎪ ⎪ dT R ⋅ T2 R ⋅ T2 ⎪⎪ ⎪⎪ ~ if C 0 ∆ = p ⎬ ⎨ ⎪ ⎪ o ⎪ ⎪K (T ) = K (T ) exp⎧ ∆H RXN (TR ) ⎛⎜ 1 − 1 ⎞⎟⎫ ⎨ ⎬ p 1 ⎟ ⎜ ⎪ ⎪ p 2 R ⎝ T1 T2 ⎠⎭ ⎩ ⎪⎭ ⎪⎩ 12 To increase the conversion in an exothermic rxn, use multiple reactors with interstate cooling: Xe Xeb For Endothermic Rxn (you need heating) Xe Optimum Feed Temperature T = To - Adiabatic Reactor of fixed size Reversible & Exothermic Rxn ∆H rx ⋅X C pA if ∆H rx < 0 as X X 2 3 350 As T Xe ;T 1 500 600 T, K but the (-rA) decreases! (so the conversion is achieved at the end of the reactor) So there must be an optimum temperature to achieve max X. Curve A: Rxn rate slow, rxn dictated by rate of rxn and reactor volume As T ,r ,X Curve B: Rxn rate very rapid. Virtual equilibrium reached in X dictated by equilibrium conversion 13 Adiabatic Rxn Algorithm Suppose 1. Choose X Calculate T Calculate k Calculate T/To Calculate CA Calculate CB Calculate KC Calculate -rA 2. Increment X and then repeat calculations. 3. When finished, plot vs. X or use some numerical technique to find V. Levenspiel Plot for an exothermic, adiabatic reaction. 14 Consider: PFR Shaded area is the volume. For an exit conversion of 40% CSTR For an exit conversion of 70% Shaded area is the reactor volume. For an exit conversion of 40% For an exit conversion of 70% We see for 40% conversion very little volume is required. CSTR+PFR 15 For an intermediate conversion of 40% and exit conversion of 70% Looks like the best arrangement is a CSTR with a 40% conversion followed by a PFR up to 70% conversion. Evaluating the Heat Exchanger Term Energy transferred between the reactor and the coolant: Assuming the temperature inside the CSTR, T, is spatially uniform: 16 At high coolant flow rates the exponential term will be small, so we can expand the exponential term as a Taylor Series, where the terms of second order or greater are neglected, then: Since the coolant flow rate is high, Ta1 Ta2 Ta: Multiple Steady States (MSS) where Heat generated term Heat removed term 17 R(T): Varying Entering Temperature R(T) Slope = Cpo (1+κ) Increase To T Vary non-adiabatic parameter κ: if you increase FA0 (molar flow rate) or decrease heat – exch area then κ will decrease R(T) ⎧⎪ Ua ⎫⎪ ⎬ ⎨κ = FA0 C p0 ⎪⎭ ⎪⎩ T + κ ⋅ Ta Tc = 0 1+ κ κ=∞ κ= 0 κ decrease Ta T To G(T) G(T) High E G(T) increasing T Low E T T For a first order rxn: X= τ ⋅k 1+τ ⋅ k 18 Ignition – Extinction Curve: The point of intersection of G(T) and R(T) give Tst-st. By plotting Tst-st vs To, we obtain ignition – extinction curve. Upper st-st As To Tst-st Lower st-st Ignition temp Extinction temp Runaway Rxns in a CSTR: R(T), G(T) Tc = T0 + κ ⋅ Ta 1+ κ ∆Trc = T * - Tc = Tc T* Tangency point R ⋅ T *2 E Reactor T if this diff. is exceeded, transition to the upper st – st will occur. At this high temp, it is undesirable or even dangerous. 19 PFR dT Ua (Ta - T) + (-rA )(-∆H Rxn (T)) = m dV ∑ Fi Cpi (for a single rxn) i =1 When q multiple rxns occur with m species: [ q dT = dV Ua (Ta - T) + ∑ (-rij ) - ∆H Rxn,ij (T) ] i : rxn j : species i =1 m ∑F C j=1 j pj Rxn 1 : Ex: k1 A ⎯⎯→ B k2 B ⎯⎯→ C dT Ua (Ta - T) + (-r1A )[- ∆H Rxn,1A (T) ] + (-r2B )[- ∆H Rxn,2B (T) ] = dV FA C pA + FB C pB + FC C pC CSTR • • Q - W s + FA0 n ∑Θ i =1 i ⋅ C pi ⋅ [T - T0 ] - [∆ H RXN (T) ]⋅ [rA V ] = 0 (for a single rxn) When q multiple rxns take place • • 0 Q - W s + FA0 q n ∑Θ i =1 UA(T a - T) + FA0 i ⋅ C pi ⋅ [T - T0 ] - V ⋅ ∑ rij ∆ H RXN, ij (T) = 0 i =1 n q i =1 i =1 ∑ Θ i ⋅ C pi ⋅ [T - T0 ] - V ⋅ ∑ rij ∆H RXN, ij (T) = 0 20