23-Chem-A2 Unit Operations and Separation Processes · December 2017
Question 4 of 6: Part B — Separation Processes
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. National Exams — 16-CHEM-A2 Unit Operations and Separation Processes, December 2017. 3 hours, open book (one text). Six problems — Part A Unit Operations (A1–A3) and Part B Separation Processes (B1–B3), each worth 25 marks; the rubric asks for at least two problems from each part (only the first two per part are marked). All six are worked below for completeness as a study resource.
Reference texts. McCabe, Smith & Harriott, Unit Operations of Chemical Engineering (7th ed.); Geankoplis, Transport Processes and Separation Process Principles (4th ed.); Coulson & Richardson, Chemical Engineering Vol. 2 (particle technology, absorption, drying, crystallization); Treybal, Mass-Transfer Operations (3rd ed.); Perry's Chemical Engineers' Handbook (9th ed.).
Notes on chart / model reads. A1's minor-loss balance uses the Blasius smooth-pipe friction factor $f=0.079\,Re^{-0.25}$ (Fanning) — valid to $Re\approx10^5$, and all three test points fall in $3.8$–$8.9\times10^4$; the 1.5 m is taken as the head lost across the pipe-and-valve test section, because also charging an exit velocity head would make the measured open-valve flow physically impossible (see A1). B1(b) reads the Sherwood/Eckert flooding line supplied with the paper at abscissa $0.061$, giving flooding ordinate $\approx0.17$ (graphical, $0.16$–$0.18$), with the tabulated packing factor $F=160\ \text{ft}^{-1}$ converted to $525\ \text{m}^{-1}$ for the SI form of the ordinate. B3(a) sizes the vessel from its starting liquor volume, allowing for the water evaporated during the batch. Engineering choices are flagged in Check callouts.
Part A — Unit Operations
Part B — Separation Processes
Question B1: Height and Diameter of a Packed Absorption Tower (25 marks)
Find. (a) the packed height $Z=H_{oy}N_{oy}$; (b) the tower diameter operating at 50% of the flooding velocity.
Figure B1 — Dilute acetone is absorbed into a falling water stream; 96% recovery with an absorption factor near 1 forces a tall, slender tower.
(a) Packed height
Approach. Convert the streams to solute-free molar flows, fix the minimum (then operating) water rate at the bottom pinch, get $N_{oy}$ from the Colburn dilute expression, and multiply by the height per transfer unit $H_{oy}$.
Molar flows and terminal compositions. The entering gas has $\bar M=0.02(58.08)+0.98(28.97)=29.55\ \text{kg/kmol}$, so $G=450/29.55=15.23\ \text{kmol/hr}$; the inert-air carrier is $G_s=0.98(15.23)=14.92\ \text{kmol/hr}$ and acetone in $=0.3046\ \text{kmol/hr}$. Removing 96% leaves $0.0122\ \text{kmol/hr}$, giving mole ratios $$Y_1=\frac{0.3046}{14.92}=0.02041,\qquad Y_2=\frac{0.0122}{14.92}=8.16\times10^{-4},\qquad y_2=8.16\times10^{-4}.$$
Minimum and operating liquid rate. The minimum solvent leaves the bottom in equilibrium with the entering gas: $x_1^{*}=y_1/m=0.02/2.5=0.008$, i.e. $X_1^{*}=0.008/(1-0.008)=0.008065$. A solute balance on the carriers (pure water, $X_2=0$) gives $$L_{s,\min}=\frac{G_s(Y_1-Y_2)}{X_1^{*}}=\frac{14.92(0.02041-0.000816)}{0.008065}=\boxed{36.3\ \text{kmol/hr}},$$ and the operating rate is $L_s=1.20\,L_{s,\min}=\boxed{43.5\ \text{kmol/hr}}$ water.
Number of transfer units (Colburn). With the absorption factor $A=\dfrac{L_s}{mG_s}=\dfrac{43.5}{2.5(14.92)}=1.166$ (so $1/A=0.857$) and $x_2=0$, $$N_{oy}=\frac{\ln\!\left[\left(1-\tfrac1A\right)\dfrac{y_1}{y_2}+\tfrac1A\right]}{1-\tfrac1A}=\frac{\ln\!\left[0.1426(24.5)+0.857\right]}{0.1426}=\boxed{10.3}.$$
Height of a transfer unit and packed height. The paper's own relation, with $mG/L=1/A=0.857$: $$H_{oy}=H_y+\frac{mG}{L}H_x=0.54+0.857(0.32)=\boxed{0.814\ \text{m}},$$ hence $$\boxed{Z=H_{oy}N_{oy}=0.814(10.3)=8.4\ \text{m}.}$$ The large $N_{oy}\approx10$ is expected — 96% removal with $A$ only modestly above 1 (a near-pinched line) demands many transfer units.
(b) Tower diameter at 50% flooding
Approach. Enter the Sherwood/Eckert flooding chart on the flow-parameter abscissa using the bottom mass rates, read the flooding ordinate, solve for the flooding gas flux (with consistent SI units), halve it, and size the area from the actual gas mass flow.
Flow-parameter abscissa. Using the bottom (largest) mass rates — gas $G'=450\ \text{kg/hr}$, liquid $L'=L_sM_w+\text{(acetone absorbed)}=43.5(18.02)+0.29(58.08)\approx801\ \text{kg/hr}$: $$\frac{L'}{G'}\sqrt{\frac{\rho_G}{\rho_L}}=\frac{801}{450}\sqrt{\frac{1.181}{998.4}}=1.78(0.0344)=0.0612.$$
Read the flooding ordinate and solve the capacity group. On the supplied chart the flooding line at abscissa $0.061$ lies between the 0.10 and 0.20 gridlines and reads $\approx0.17$. The ordinate is $\dfrac{(G'_f)^2 F\Phi\mu_L^{0.1}}{\rho_G\rho_L g}$, which is dimensionless only when $F$ carries reciprocal length in the same units as $g$. The paper's $F=160$ is the tabulated value for 2.54-cm ceramic Raschig rings in ft$^{-1}$, so in SI $F=160/0.3048=525\ \text{m}^{-1}$. With $\Phi=1.0$ and $\mu_L^{0.1}=0.86^{0.1}=0.985$: $$G'_f=\sqrt{\frac{0.17\,\rho_G\rho_L g}{F\Phi\mu_L^{0.1}}}=\sqrt{\frac{0.17(1.181)(998.4)(9.81)}{525(1.0)(0.985)}}=\boxed{1.95\ \text{kg/m}^2\!\cdot\!\text{s}}.$$
Design flux and diameter. At 50% of flooding the design gas flux is $G'_\text{op}=0.5(1.95)=0.975\ \text{kg/m}^2\!\cdot\!\text{s}$. With the actual gas mass flow $450\ \text{kg/hr}=0.125\ \text{kg/s}$, $$A=\frac{0.125}{0.975}=0.128\ \text{m}^2\ \Longrightarrow\ D=\sqrt{\frac{4A}{\pi}}=\boxed{0.40\ \text{m}}.$$ The tower is therefore $\approx0.40$ m (400 mm) in diameter and 8.4 m tall — the slender, tall geometry typical of a dilute-gas absorber at high recovery.
Quantity
Value
Operating water rate $L_s$
43.5 kmol/hr ($1.2\times$ min)
Absorption factor $A$ / $N_{oy}$
1.166 / 10.3
Height per transfer unit $H_{oy}$
0.814 m
Packed height $Z$
≈8.4 m
Flooding gas flux $G'_f$
1.95 kg/m²·s
Tower diameter $D$ (50% flood)
≈0.40 m
Check: the flooding ordinate is read graphically ($0.16$–$0.18$), so $D\approx0.40$–$0.41$ m. Entering $F=160$ as if it were m$^{-1}$ would over-state the flooding flux by $\sqrt{3.28}$ and under-size the diameter by about 26%; the liquid load $L'$ is taken at the bottom where both phase loads are largest (the conservative sizing point).