23-Chem-A4 Chemical Reactor Engineering · Undated paper
Question 4 of 5: Effectiveness Factor from Particle-Size Experiments
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
National Exams / EGBC — 16-Chem-A4, Chemical Reactor Engineering, May 2019 (archived without a date as “undated (16-Chem-A4)”) — 23-Chem-A4 Chemical Reactor Engineering. Three-hour open-book exam: one textbook of the candidate's choice, personal unit-conversion / mathematical tables (e.g. CRC Handbook) and any non-communicating calculator. Five questions are printed and any four constitute a complete paper (25 points each); all five are solved below. No credit is given for deriving rate expressions or standard formulas available in the textbook, so the design equations are quoted with their source and applied.
Reference texts: H. S. Fogler, Elements of Chemical Reaction Engineering (5th ed., Pearson) — PFR design equation, Langmuir–Hinshelwood rate-law analysis, residence-time distributions, internal effectiveness factor and external mass transfer in packed beds; O. Levenspiel, Chemical Reaction Engineering (3rd ed., Wiley) — tracer (pulse) experiments and film/pore-diffusion diagnostics for solid-catalysed reactions.
Question 4: Effectiveness Factor from Particle-Size Experiments (25 marks)
Given. First-order reaction, isothermal porous spheres, constant $D_{A,eff}$; $k$ and $D_{A,eff}$ independent of size; no external resistance.
Particle radius $R$ (mm)
2
0.5
0.15
Measured rate (mol A/L cat·s)
2.5
8.9
20.1
Find. The internal effectiveness factor $\eta$ for each particle radius.
Figure 4 — In a large pellet the reactant is used up before it reaches the core ($\eta\ll1$). Even the smallest pellet tested still has a noticeable internal gradient ($\eta\approx0.76$).
Approach. None of the pellets can be assumed free of diffusion limits; the rate is still rising steeply as the radius falls. So the intrinsic rate is unknown. For a first-order sphere the observed rate is $r_{obs}=\eta(\phi)\,r_{int}$ with $\eta=\dfrac{3}{\phi^2}(\phi\coth\phi-1)$ and $\phi=R\sqrt{k\rho_c/D_{A,eff}}$ (Fogler Ch. 15). Since $k$ and $D_{A,eff}$ do not depend on size, $\phi\propto R$. That leaves two unknowns, the ratio $\phi/R$ and $r_{int}$. A ratio of two rates removes $r_{int}$ and fixes $\phi/R$, and the third rate is a check.
Size ratio removes the intrinsic rate. For radii $R_1$ and $R_2$: $\dfrac{r_1}{r_2}=\dfrac{\eta(\phi_1)}{\eta(\phi_2)}$ with $\phi_1/\phi_2=R_1/R_2$.
Pair 2 mm / 0.5 mm. $\eta(4\phi_{0.5})/\eta(\phi_{0.5})=2.5/8.9=0.281$. Solving numerically gives $\phi/R=14.1\ \text{mm}^{-1}$. In the strong-diffusion limit $\eta\approx3/\phi$ the ratio would be exactly 0.25, so the 0.5 mm pellet is only partly limited.
Pair 0.5 mm / 0.15 mm (check). $\eta(\phi_{0.5})/\eta(0.3\phi_{0.5})=8.9/20.1=0.443$, which gives $\phi/R=15.6\ \text{mm}^{-1}$. The two estimates agree within 10%, so the data are consistent with a single Thiele modulus.
Best fit to all three rates. Minimizing the relative error over the three points gives $\phi/R=15.3\ \text{mm}^{-1}$ and $r_{int}=26.3\ \text{mol/L}\cdot\text{s}$. The Thiele moduli are $\phi=30.6,\ 7.66,\ 2.30$, and the fitted model reproduces all three measured rates within 1%.
Effectiveness factors. $$\boxed{\eta_{2\,\text{mm}}=0.095,\qquad \eta_{0.5\,\text{mm}}=0.34,\qquad \eta_{0.15\,\text{mm}}=0.76}$$ As a hand check on the largest pellet, the strong-diffusion asymptote $\eta\approx3/\phi=3/30.6=0.098$ agrees with the exact value.
The physical reading: the 2 mm pellet does less than a tenth of the work its catalyst could do. Even the 0.15 mm powder loses about a quarter of its activity to pore diffusion. An intrinsic rate of about 26 mol/L·s would only be reached with particles well below 0.1 mm radius ($\phi<1$ requires $R<0.065$ mm).
Check
A tempting shortcut is to call the smallest particle diffusion-free and divide by 20.1. That gives $\eta=0.124,\ 0.443,\ 1.00$, which is inconsistent with the data: if $\eta_{0.15}=1$ then $\phi_{0.15}<1$, so $\phi_{0.5}<3.3$ and $\eta_{0.5}$ would have to exceed 0.6, not 0.44. The simultaneous fit above removes that contradiction.