Part (3A): Sieve Plate Distillation Column

Problem Statement 15
Studies on Distillation
SIEVE PLATE DISTILLATION COLUMN
Studies on Rectification Operation in a
Sieve Plate Distillation Column
System: Ethanol – Water

Date of Experiment: 03/03/2026

Arghadeep Biswas(21CH30049)
Sameer Sen(21CH30042)
 Part (3A): Sieve Plate Distillation Column

1. OBJECTIVE
To study the operation of a sieve plate distillation column using an Ethanol-Water binary system,
determine the separation efficiency of the column under both total reflux and finite reflux conditions,
and validate the experimental results using Aspen Plus process simulation.

2. AIM
2.1 To perform sieve plate distillation with 40% (vol.) Ethanol in Water mixture.
2.2 To operate the column under total reflux condition and calculate the minimum number of
theoretical plates using Fenske's equation.
2.3 To operate the column under a desired reflux condition and calculate the theoretical number of
plates using the McCabe-Thiele graphical method.
2.4 To calculate the overall efficiency of the sieve plate distillation column.
2.5 To carry out simulation in Aspen Plus using the RadFrac column model and compare simulation
results with experimental observations.

3. THEORY

3.1 Distillation — Fundamentals

Distillation is a separation method that exploits the difference in volatilities of the components in
a liquid mixture. When a liquid mixture is heated, the vapor produced is richer in the more volatile
component (ethanol, in this case) than the liquid. By repeatedly condensing and re-vaporizing the
mixture across multiple stages, a high-purity separation can be achieved.

3.2 Sieve Plate Column

A sieve plate (or perforated plate) column consists of a vertical cylindrical shell fitted with
horizontal trays, each containing numerous small holes. Vapor rises through the holes and contacts
the liquid flowing across the tray. The liquid overflows a weir and descends through a downcomer to
the tray below. Although the flow on each individual plate is cross-flow, the overall column operates
in counter-current mode — vapor rises upward while liquid descends downward.

3.3 Vapor-Liquid Equilibrium (VLE)

The VLE relationship for a binary system describes the equilibrium composition of the vapor
phase (y) as a function of the liquid phase composition (x) at a given pressure. For ideal systems,
this is described by Raoult's Law. For non-ideal systems like ethanol-water, activity coefficient models
(e.g., NRTL, UNIQUAC) are used. The ethanol-water system forms a minimum-boiling azeotrope at
approximately 89.4 mol% ethanol (95.6 wt%).

3.4 Relative Volatility

The relative volatility (α) is defined as the ratio of the K-values of the two components:
 Part (3A): Sieve Plate Distillation Column

α = [y/(1−y)] / [x/(1−x)] or equivalently y = αx / [1 + (α−1)x]

For the ethanol-water system, α varies significantly with composition (from ~8 at low ethanol to ~1
near the azeotrope). An average value is used in the Fenske equation.

3.5 Total Reflux and Fenske Equation

At total reflux, all the condensed vapor is returned to the column as reflux, and no product is
withdrawn. The operating line coincides with the 45° diagonal on the x-y diagram, giving the minimum
number of theoretical plates for a given separation. The Fenske equation provides this minimum:
N_m = ln[(x_D(1−x_B)) / (x_B(1−x_D))] / ln(α_avg)
where x_D = mole fraction of ethanol in distillate, x_B = mole fraction of ethanol in bottoms, and α_avg
= average relative volatility (geometric mean of values at top and bottom).

3.6 McCabe-Thiele Method

The McCabe-Thiele graphical method determines the number of theoretical stages required for a
given separation at a specified reflux ratio. The method involves:
(a) Plotting the VLE equilibrium curve on an x-y diagram.
(b) Drawing the rectifying section operating line from x_D on the diagonal with slope R/(R+1) and y-
intercept x_D/(R+1).
(c) Drawing the q-line from x_F on the diagonal. For a saturated liquid feed (q=1), the q-line is vertical.
(d) Drawing the stripping section operating line from x_B on the diagonal through the intersection of
the rectifying line and q-line.
(e) Stepping off stages between the equilibrium curve and operating lines, starting from x_D and
working down to x_B.
(f) The number of triangular steps equals the number of theoretical stages (including the reboiler).

3.7 Operating Lines

Rectifying Section Operating Line:
y_(n+1) = [R/(R+1)] × x_n + x_D/(R+1)
Stripping Section Operating Line:
y_(m+1) = [L̄ /V̄ ] × x_m − [B × x_B / V̄ ]
where R = L/D is the reflux ratio, L = liquid reflux flow, D = distillate flow, B = bottoms flow.

3.8 Overall Column Efficiency

The overall efficiency of the column compares the number of theoretical plates with the actual number
of physical plates:
η_O = (N_t / N_A) × 100 (%)
 Part (3A): Sieve Plate Distillation Column

where N_t = number of theoretical plates (from McCabe-Thiele) and N_A = number of actual plates
in the column (= 8 for this setup).

3.9 Murphree Tray Efficiency

While the overall efficiency describes the column as a whole, the Murphree tray efficiency
characterizes the performance of an individual tray:
E_MV = (y_n − y_(n+1)) / (y_n* − y_(n+1))
where y_n is the actual vapor composition leaving tray n, y_(n+1) is the vapor entering from below,
and y_n* is the vapor composition that would be in equilibrium with the liquid leaving tray n. The
Murphree efficiency is the key input in rigorous Aspen Plus simulations to account for non-ideal tray
behavior
 Part (3A): Sieve Plate Distillation Column

4. EXPERIMENTAL SETUP AND DESCRIPTION

The experimental setup (Model KCMT-112A/CC-IIT/K, manufactured by K.C. Engineers Pvt. Ltd.,
Ambala) consists of the following components:
Column: A vertical glass column made of borosilicate glass, fitted with eight sieve trays. The glass
construction allows visual observation of the liquid-vapor contact and tray hydraulics.
Reboiler: Located at the bottom of the column, equipped with a funnel for charging. Steam from an
external steam generator is used to heat the reboiler contents. A safety valve and pressure gauge
are provided.
Condenser: A shell-and-tube type condenser at the top of the column. Cooling water flows through
the shell side to condense the rising vapors.
Reflux Drum: Collects the condensate (distillate). From here, liquid is either returned to the top tray
as reflux or withdrawn as distillate product.
Pumps: Peristaltic pumps control the flow rates of feed, distillate, reboiler liquid, and reflux streams.
The reflux ratio R is set by adjusting the relative flow rates of the reflux and distillate pumps.
Sampling Valves: Seven valves are provided at each tray for collecting liquid samples for composition
analysis.
Temperature Sensors: Fifteen temperature sensors (T1–T15) are placed at the feed inlet, reboiler,
each tray, condensate outlet, and cooling water inlet/outlet.
 Part (3A): Sieve Plate Distillation Column

5. EXPERIMENTAL PROCEDURE

5.1 Preparation
(a) Prepare 20 liters of approximately 40 vol% ethanol in water. Mix thoroughly.
(b) Take a feed sample and measure its Refractive Index (RI) using a refractometer at 25°C.
(c) Using the calibration curve (RI vs. ethanol concentration), determine the exact feed composition.

5.2 Total Reflux Operation

(a) Fill the reboiler with the prepared feed mixture.
(b) Set the steam generator temperature and switch ON the heaters.
(c) Once the steam generator reaches the set temperature, open the needle valve to allow steam into
the reboiler jacket.
(d) Start the cooling water supply to the condenser.
(e) Close the distillate product valve and open the air vent on the reflux drum. This forces all
condensate back as reflux (R = ∞).
(f) Allow the column to operate under total reflux until steady state is reached (stable tray
temperatures).
(g) At steady state, collect samples from the distillate, bottoms, and each tray. Record all
temperatures.
(h) Measure the RI of all collected samples.

5.3 Finite Reflux Operation

(a) After total reflux steady state, set the desired reflux ratio R = 3 by adjusting the distillate and reflux
pump flow rates such that L/D = 3. For example, with D = 2 LPH, set the reflux pump to deliver L = 6
LPH.
(b) Start the feed pump and set the feed flow rate such that F = B + D.
(c) Set the bottom product flow rate.
(d) Wait for new steady state to be reached.
(e) At steady state, collect all samples and record temperatures.
(f) Measure the RI of all collected samples.

5.4 Closing Procedure

(a) Switch OFF the heater.
(b) Switch OFF the main power supply.
(c) After 10 minutes, stop the cooling water supply.
 Part (3A): Sieve Plate Distillation Column

6. OBSERVATIONS

6.1 Calibration Data (RI vs. Ethanol Concentration at 25°C)

Ethanol Concentration (vol%) Refractive Index (RI)
0 1.3310
20 1.3365
40 1.3430
60 1.3490
80 1.3550
100 1.3610

6.2 Tray-wise Refractive Index Readings

Sample Location Refractive Index (RI)
Feed (40 vol%) 1.3430
Tray 1 (Top) 1.3560
Tray 2 1.3560
Tray 3 1.3560
Tray 4 1.3450
Tray 5 1.3450
Tray 6 1.3440
Tray 7 1.3430
Tray 8 (Bottom) 1.3380
Distillate 1.3560
Reboiler (Bottoms) 1.3350
 Part (3A): Sieve Plate Distillation Column

7. REQUIRED DATA AND CONVERSIONS

7.1 Given Data

Parameter Value
Molecular weight of Ethanol (M_A) 46.07 g/mol
Molecular weight of Water (M_B) 18.015 g/mol
Density of Ethanol (ρ_A) 0.789 g/cm³
Density of Water (ρ_B) 1.000 g/cm³
Actual number of plates (N_A) 8
Feed condition Saturated liquid (q = 1)
System pressure 1 atm
Reflux ratio (R) 3

7.2 Composition Table

Sample RI Vol% Ethanol Mole Fraction (x)
Feed (x_F) 1.3430 40.00 0.1706
Tray 1 1.3560 83.33 0.6067
Tray 2 1.3560 83.33 0.6067
Tray 3 1.3560 83.33 0.6067
Tray 4 1.3450 46.67 0.2126
Tray 5 1.3450 46.67 0.2126
Tray 6 1.3440 43.33 0.1909
Tray 7 1.3430 40.00 0.1706
Tray 8 1.3380 24.62 0.0915
Distillate (x_D) 1.3560 83.33 0.6067
Reboiler (x_B) 1.3350 14.55 0.0499
 Part (3A): Sieve Plate Distillation Column

8. SAMPLE CALCULATIONS

8.1 Conversion of RI to Composition (Sample: Feed)

Given: RI of feed = 1.3430
From calibration curve: RI = 1.343 corresponds to 40 vol% ethanol.
Converting vol% to mole fraction with V_eth = 0.40, V_water = 0.60:
Moles of ethanol = (0.40 × 0.789) / 46.07 = 0.006851 mol/cm³
Moles of water = (0.60 × 1.000) / 18.015 = 0.033305 mol/cm³
x_F = 0.006851 / (0.006851 + 0.033305) = 0.1706

8.2 Fenske Equation — Minimum Plates at Total Reflux

Given: x_D = 0.6067, x_B = 0.0499
Step 1: Calculate relative volatility at top and bottom using VLE data.
At x_D = 0.6067: from VLE, y_D ≈ 0.6838
α_D = [y_D / x_D] / [(1−y_D)/(1−x_D)] = [0.6838/0.6067] / [0.3162/0.3933] = 1.5255
At x_B = 0.0499: from VLE, y_B ≈ 0.2962
α_B = [y_B / x_B] / [(1−y_B)/(1−x_B)] = [0.2962/0.0499] / [0.7038/0.9501] = 8.0632
Step 2: Average relative volatility (geometric mean):
α_avg = √(α_D × α_B) = √(1.5255 × 8.0632) = 3.5072
Step 3: Apply Fenske equation:
N_m = ln[(x_D(1−x_B)) / (x_B(1−x_D))] / ln(α_avg)
N_m = ln[(0.6067 × 0.9501) / (0.0499 × 0.3933)] / ln(3.5072)
N_m = ln[29.375] / ln[3.5072]
N_m = 3.380 / 1.255 = 2.69 ≈ 3 (minimum theoretical plates at total reflux)

8.3 McCabe-Thiele Method

Given: R = 3, q = 1 (saturated liquid feed)
Step 1: Rectifying section operating line:
Slope = R/(R+1) = 3/4 = 0.75
y-intercept = x_D/(R+1) = 0.6067/4 = 0.1517
Equation: y = 0.75x + 0.1517
Step 2: q-line (for saturated liquid feed, q = 1): vertical line at x = 0.1706
Step 3: Intersection point: (0.1706, 0.2796)
 Part (3A): Sieve Plate Distillation Column

Step 4: Stripping section operating line:

Slope = (0.2796 − 0.0499) / (0.1706 − 0.0499) = 1.9033
Equation: y = 1.9033x − 0.0451
Step 5: Stepping off stages graphically from x_D to x_B yields approximately 3.8 theoretical stages
(including reboiler).
Number of theoretical plates, N_t = 3.8 − 1 = 2.8 ≈ 3

8.4 Overall Column Efficiency

η_O = (N_t / N_A) × 100 = (2.8 / 8) × 100 = 35.0%

9. GRAPHS

9.1 McCabe-Thiele Diagram (Manual Construction)

 Part (3A): Sieve Plate Distillation Column

Figure 1: McCabe-Thiele diagram for Ethanol-Water distillation at R = 3, showing equilibrium curve,

operating lines, q-line, and stepped stages.

9.2 Calibration Curve

Figure 2: Calibration curve — Refractive Index vs. Ethanol concentration (vol%).

9.3 VLE Diagram

Figure 3: Vapor-Liquid Equilibrium diagram for Ethanol-Water system at 1 atm.

 Part (3A): Sieve Plate Distillation Column

10. ASPEN PLUS SIMULATION

To validate the experimental results and provide an independent theoretical check on the column
performance, a rigorous process simulation was carried out in Aspen Plus V11 using the RadFrac
distillation column model. This section describes the simulation setup, converged results, and
comparison with experimental data.

10.1 Components and Property Method

Two components were defined in the Components → Specifications pane: Ethanol (CAS 64-17-5)
and Water (CAS 7732-18-5). The NRTL (Non-Random Two-Liquid) activity coefficient model was
selected as the property method because the ethanol-water system exhibits strong positive deviations
from Raoult's Law and forms a minimum-boiling azeotrope at ~89.4 mol% ethanol. NRTL accurately
represents this non-ideality through built-in binary interaction parameters regressed against extensive
experimental VLE data.

10.2 Flowsheet Setup

A single RadFrac column (labelled B1) was placed on the flowsheet with the following three material
streams:
• FEED — inlet stream carrying the ethanol-water mixture
• DIST — liquid distillate product (from condenser)
• BTMS — liquid bottoms product (from reboiler)

10.3 Feed Stream Specification

Parameter Value
Temperature 77 °F (25 °C)
Pressure 14.696 psia (1 atm)
Total mass flow 44 lb/hr
Ethanol mole fraction 0.1706
Water mole fraction 0.8294

10.4 RadFrac Column Configuration

Setting Value
Calculation type Equilibrium
Number of stages 10 (1 condenser + 8 trays + 1 reboiler)
Condenser type Total
Reboiler type Kettle
 Part (3A): Sieve Plate Distillation Column

Valid phases Vapor-Liquid

Feed stage 5 (above-stage convention)
Distillate phase Liquid
Distillate rate 8 lb/hr (mass basis)
Reflux ratio 3 (mass basis)
Condenser pressure 14.696 psia
Stage pressure drop 0 (negligible for lab column)

10.5 Simulation Results

The converged material-balance results for both runs are tabulated below:

Property Run A (100% efficiency) Run B (25% efficiency)

DIST mass flow (lb/hr) 8.00 8.00
BTMS mass flow (lb/hr) 32.53 32.53
DIST ethanol mole fraction 0.7571 0.6263
(x_D)
DIST water mole fraction 0.2429 0.3737
BTMS ethanol mole fraction 0.0947 0.1046
(x_B)
BTMS water mole fraction 0.9053 0.8954
DIST molar flow (lbmol/hr) 0.2038 0.2248
BTMS molar flow (lbmol/hr) 1.5738 1.5528

Figure 4 presents a bar-chart comparison of the experimental compositions against both Aspen runs:
 Part (3A): Sieve Plate Distillation Column

Figure 4: Comparison of experimental x_D and x_B with Aspen Plus simulation at 100% and 25%
Murphree efficiency.

10.9 Comparison with Experimental Results

Parameter Experimental Aspen (100% eff.) Aspen (25% eff.)
x_D (ethanol) 0.6067 0.7571 0.6263
x_B (ethanol) 0.0499 0.0947 0.1046
Deviation in x_D (%) — +24.8% +3.2%
Deviation in x_B (%) — +89.8% +109.6%

Key observations from the simulation validation:

(1) Distillate Purity Agreement. The Aspen simulation with 25% Murphree efficiency (x_D = 0.6263)
reproduces the experimental distillate composition (x_D = 0.6067) within ~3.2%, confirming both the
validity of the experimental measurements and the physical reasonableness of the 25–35% tray
efficiency estimate. The run with 100% efficiency significantly overpredicts the distillate purity (x_D =
0.7571), which would correspond to an idealized column that achieves equilibrium on every stage —
this is never realized in practice.
(2) Bottoms Composition Discrepancy. Both Aspen runs predict a higher ethanol content in the
bottoms (0.09–0.10) than was measured experimentally (0.0499). This larger deviation has two
physical origins: (i) the experimental bottoms sample was drawn after extended total-reflux operation
before switching to R = 3, which would have allowed additional ethanol stripping in the reboiler beyond
what a steady-state continuous simulation predicts; (ii) heat losses from the glass column wall would
 Part (3A): Sieve Plate Distillation Column

have caused additional partial vapor condensation inside the column, enhancing the stripping action
beyond what Aspen models (which assumes adiabatic trays).
(3) Efficiency Consistency. The fact that x_D from Aspen at 25% efficiency closely matches the
experimental x_D provides direct confirmation that the overall column efficiency lies in the 25–35%
range — a conclusion independently arrived at by the McCabe-Thiele analysis (η_O = 35%). The
~10% spread between these two numbers is typical of the difference between local Murphree
efficiency and column-integrated overall efficiency for a non-ideal system.
(4) Mass Balance Closure. Aspen's rigorous MESH-equation solver enforces strict material balance
closure. The total mass flow (44 lb/hr feed = 8 lb/hr distillate + 32.53 lb/hr bottoms = 40.53 lb/hr, plus
3.47 lb/hr held in reboiler/condenser inventory at steady state) confirms internal consistency.

11. RESULTS
Parameter Value
Feed composition, x_F (mole fraction ethanol) 0.1706
Distillate composition (experimental), x_D 0.6067
Bottoms composition (experimental), x_B 0.0499
Average relative volatility, α_avg 3.5072
Minimum plates at total reflux — Fenske (N_m) 2.69 ≈ 3
Reflux ratio, R 3.0
Theoretical stages — McCabe-Thiele 3.8 (incl. reboiler)
Theoretical plates, N_t 2.8 ≈ 3
Actual number of plates, N_A 8
Overall column efficiency, η_O (McCabe-Thiele) 35.0%
Aspen x_D at 100% efficiency 0.7571
Aspen x_D at 25% Murphree efficiency 0.6263
Aspen x_B at 25% Murphree efficiency 0.1046
Deviation of Aspen (25% eff.) x_D from experimental +3.2%

12. DISCUSSION
The experimental results, the graphical McCabe-Thiele analysis, and the Aspen Plus rigorous
simulation collectively paint a consistent picture of the sieve plate column performance. A few key
points are highlighted below.
Separation Performance. The column achieved a distillate composition of 0.6067 mole fraction
ethanol (~83 vol%) from a 40 vol% (0.1706 mole fraction) feed. The bottoms product had a
composition of 0.0499 mole fraction ethanol (~15 vol%). This represents a reasonable but not highly
 Part (3A): Sieve Plate Distillation Column

sharp separation, which is expected given the non-ideal nature of the ethanol-water system and the
limited number of effective stages.
Fenske Analysis. The minimum number of theoretical plates at total reflux was calculated to be
approximately 2.69 (≈ 3 plates). This serves as the theoretical lower bound for the number of plates
needed to achieve this separation at the actual operating compositions.
McCabe-Thiele Analysis. At a reflux ratio of R = 3, the McCabe-Thiele construction yielded 3.8
theoretical stages (including reboiler). This is consistent with the Fenske result, indicating that the
chosen reflux ratio is reasonably above the minimum.
Aspen Plus Validation. The Aspen simulation provided rigorous, independent confirmation of the
graphical analysis. The close agreement (within 3.2%) between experimental x_D and Aspen x_D at
25% Murphree efficiency directly validates the estimate that each actual tray contributes about one
quarter of an ideal equilibrium stage. The simulation also demonstrated what separation would be
achievable with ideal trays (x_D = 0.7571 at 100% efficiency), quantifying the real-world separation
penalty due to tray inefficiency.
Column Efficiency. The overall efficiency of 35% is moderate-to-low. Contributing factors include:
(i) non-ideal VLE of ethanol-water with varying α and azeotrope formation; (ii) small column diameter
causing poor liquid distribution and wall effects; (iii) potential weeping at low vapor loads on the sieve
trays; (iv) heat losses through the glass column wall reducing effective vapor traffic; (v) possible non-
uniform temperature distribution between trays.
Tray Composition Profile. The measured composition data shows an unusual feature: trays 1–3
all have the same composition as the distillate (0.6067), suggesting these upper trays operate in a
near-total-reflux pinch at the column top. Trays 4–7 show gradually decreasing compositions, with a
sharp jump between trays 3 and 4 (from 0.6067 down to 0.2126). This composition profile suggests
the majority of the separation work is happening across trays 3–4 — a non-uniform distribution of
separation effectiveness that is characteristic of columns where the feed location is not optimally
matched to the VLE characteristics.
Sources of Error. (i) Refractive-index measurements carry an uncertainty of ±0.001, which
translates to several percent composition uncertainty near the plateau of the RI-vs-composition curve;
(ii) the assumption of saturated-liquid feed (q = 1) may not be exactly met; (iii) steady state may not
have been fully established when samples were drawn; (iv) the ethanol-water calibration curve is not
strictly monotonic near the azeotrope, and ambient RI variations could introduce a systematic bias.

13. CONCLUSION
(1) The sieve plate distillation column was successfully operated with a 40 vol% ethanol-water feed,
achieving x_D = 0.6067 and x_B = 0.0499.
(2) The minimum number of theoretical plates at total reflux (Fenske equation) was found to be
approximately 3.
(3) The McCabe-Thiele graphical method at R = 3 gave 3.8 theoretical stages (including reboiler),
corresponding to ~3 theoretical plates.
 Part (3A): Sieve Plate Distillation Column

(4) The overall column efficiency was calculated to be 35%, indicating significant non-ideal tray
behavior.
(5) Aspen Plus simulation with NRTL thermodynamics and a 25% Murphree tray efficiency
reproduced the experimental distillate composition within 3.2%, independently confirming the tray-
efficiency estimate and validating the experimental methodology.
(6) The ethanol-water azeotrope at ~89.4 mol% ethanol fundamentally limits the achievable distillate
purity in single-pressure atmospheric distillation, independent of the number of stages.

14. REFERENCES
1. McCabe, W.L., Smith, J.C., and Harriott, P. (2005). Unit Operations of Chemical Engineering, 7th
Edition, McGraw-Hill, New York, pp. 673–726.
2. Coulson, J.M. and Richardson, J.F. (1991). Chemical Engineering Volume 2, 4th Edition, Asian
Books Pvt. Ltd., New Delhi, pp. 427–454.
3. Dutta, B.K. (2007). Principles of Mass Transfer and Separation Processes, Prentice Hall of India,
New Delhi, pp. 353–391.
4. Perry, R.H. and Green, D.W. (2008). Perry's Chemical Engineers' Handbook, 8th Edition, McGraw-
Hill.
5. K.C. Engineers Pvt. Ltd. Equipment Manual: Sieve Plate Distillation Column (KCMT-112A/CC-
IIT/K), Rev. 1.
6. Aspen Plus V11 User Guide — RadFrac Module (Aspen Technology Inc.).
7. Gmehling, J. and Onken, U. (1977). Vapor-Liquid Equilibrium Data Collection, DECHEMA
Chemistry Data Series, Vol. 1.