Experiment on Distillation Principles

Abstract

The general objective of this experiment is to investigate and understand distillation principles, the parameters affecting the operation of distillation columns and how to determine optimal operating conditions. To achieve this, two experiments were carried out. Experiment one was carried out to investigate the relationship between the column pressure drop and the boil-up rate, the second experiment was performed to determine the composition of the mixture of dichloromethane-trichloroethylene. The data obtained in experiment one and two were used to determine the overall column efficiency.

In order to investigate the pressure drop of the column, the power was set to 0.65 kW, 0.75 kW, 0.85, and lastly 0.95 kW. For each power input, the sample was collected for 10 seconds and then the procedures were repeated for each increment. Afterward, a graph of pressure drops against the boil-up rate (log/log) to determine the relationship between the two parameters. It was observed that the pressure drop and the boil- up rate increased as the power input increased and also that the degree of foaming increases as the power was increased from gentle at 0.65 kW to violent over the whole tray at 0.95 kW. The samples collected were tested for its Refractive Index. It was observed that the degree of foaming increases as the power was increased from gentle at 0.65 kW to violent over the whole tray at 0.95 kW.

The overall efficiency of the column for a power of 0.65 kW is 42.5%. However, that may not be the most optimal condition since it was not possible to test for other three power input due to some systematic and technical errors.

Table of Contents

Nomenclature…………………………………………………………………………..

Introduction…………………………………………………………………………….

Objectives………………………………………………………………………………

Methodology………………………………………………………………………………..

Results…………………………………………………….………………………………..

Discussion of Results…………..………………………………..………………………..

Conclusions ………………………………………………………………………………..

References ………………………………………………………………………………..

Nomenclature

n = number of theoretical plates

X_A = mole fraction of the more volatile component

X_B = mole fraction of the least volatile component

α_av = average relative volatility

Subscripts D, B or d, b indicates distillate and bottoms respectively

∆H̅_vap = average latent heat of vaporisation of DCM-TCE mixture (J/mol)

R = 8.314 J/mol*K

T_b(TCE) = boiling point of trichloroethylene (K)  

T_b(DCM) = boiling point of dichloromethane (K)

Introduction

Distillation is defined as a process in which a liquid or vapour mixture being made of two or more components is separated into its component fractions with the desired purity, by the input and removal of heat.

Distillation is one of the most common liquid-liquid separation processes and can be carried out in a continuous or batch system. The basic theory behind them is very simple and relies on separating a mixture being made of two or more components of different boiling points, though partial vaporisation of a liquid mixture or by partial condensation of the gas mixture. As the mixture is fed to the column, some fractions may vaporise and move up the column. The vapour components will condense and leave the column at different levels as the temperature is lower at the top of the tower. Based on a binary mixture, the more volatile component will leave at the top of the tower, and the less volatile component will leave at the bottom as a liquid.

Objectives

The objectives of this experiment are:

1. To investigate the pressure drop of the distillation column for four boil-up rates, and observe the degree of forming for each power supply
2. Use of refractometer to determine the overhead and bottom mixture composition
3. To determine the overall column efficiency

Literature Review

Distillation columns are usually made up of a vertical tower containing a series of plates. As liquid runs down the tower vapour goes towards the top. In order to understand its working principle, consider what happens when heating a liquid. At its boiling point, the molecules of the liquid possess enough kinetic energy to escape into the vapour phase (evaporation) and if the temperature decreases some molecule in the vapour phase return to the liquid phase (condensation). The same with the mixture of dichloromethane and trichloroethylene for this experiment. As heat is applied to the column, the eventually the most volatile component (in this case trichloroethylene) begins to vaporize. As the trichloroethylene vaporizes it takes with it some part of dichloromethane. The vapour mixture is then condensed and evaporated again, giving a higher mole fraction of the least volatile component in the liquid phase and a higher mole fraction of the most volatile in the vapour phase.

For this experiment, the column was be set to operate at total reflux. Which means that all the overhead products are be condensed and fed back into the top of the tower and allowed to flow to the bottom of the column, i.e. overall no top product is taken out of the system while the column is operational.

The total pressure drop across each tray is the sum of that caused by the restriction of the holes in the sieve tray, and that caused by passing through the liquid (foam) on top of the tray.

As the velocity of the vapours passing up the column increases then so does the overall pressure drop. The velocity can be monitored by varying the boil-up rate which is done by changing the power input to the reboiler. Under certain conditions where only the vapour phase is present, the trays will act as an orifice and in that, the velocity will be directly proportional to the square root of pressure drop. However; this relationship does not become visible until the head of liquid has been overcome and foaming is taking place. In a graph of log pressure drop vs. log boil-up rate, at low boil-up rates, the pressure drop will remain almost constant until foaming occurs when the pressure drop would be expected to rise sharply for unit increases in boil-up rate.

Key Definitions

Column efficiency:  The overall efficiency is defined as the ratio of the number of theoretical trays to the actual number of trays required for an entire column.

Foaming: Foaming regarding distillation column is defined as the expansion of liquid due to the passage of vapour, or gas. Although it provides good liquid-vapour contact interfacial, however, excessive foaming may lead to liquid build-up on column trays.

Boil- up Rate: Also called the distillation load. It is the rate at which the mixture is being distilled in the column.

McCabe Thiele Diagram. It is a diagram where y is plotted as a function x along the column provides an insightful graphical solution to the combined components. It is mainly used to determine the minimum theoretical plates required in a distillation column for separation of binary mixtures.

The Fenske Method: similar to McCabe Thiele Diagram, it is a method used to determine the minimum number of theoretical plates required in a distillation column for separation of binary mixtures. It uses equation (4).

Methodology

APPARATUS

       Figure 1: Distillation Column Apparatus

1. Distillation Column

1. Condenser
2. Electromagnet (reflux control)
3. Reboiler
4. Cooler
5. Bottom
6. Distillate
7. Feed

• Distillation column
• Dichloromethane 4.15L
• Trichloroethylene 5.85L
• Automatic digital refractometer
• Distilled water
• Measuring cylinder
• Conical flask
• Dropper
• Manometer
• Stopwatch

• Procedure for Experiment A: Variation of column pressure drop
  1. Ensure all the valves on the equipment are closed and then open valve 10 (V10)
  2. Reboiler heater power is switched on at the console and power to the heater is adjusted until a reading of 0.65 kW is obtained on the digital wattmeter. Water in the reboiler began to heat up and this is observed by selecting T9 (the reboiler temperature) on the process temperature digital display.
  3. The temperature is let stabilize for 5-10 minutes.
  4. Open V6 and V7 and measure the pressure difference in the manometer. Then close V6 and V7.
  5. Volume collection. Open V3 so that all condensate is delivered into a measuring cylinder for 10 seconds
  6. Few drops of the sample are taken and the refractive index for the sample is checked by using the refractometer.
  7. Repeat steps 1 to 6 for power of 0.75, 0.85 and 0.95kW.
• Procedure for Experiment B:  Determining the Mixture Composition
  1.  Using the reflectometer, the refractive index (R.I) of pure dichloromethane and pure trichloroethylene are measured.
  2. Measure the refractive index of small quantities of 25%, 50%, 75 and 100 mol percent of dichloromethane/trichloroethylene. The volume of constituents is calculated are shown in the results section.
• Procedure for Experiment C:  Overall Column Efficiency

Note: The overall efficiency was determined using the data from part A and B

Results

Experiment A:

Table 1.  Measured and calculated parameters

Table 1.  Measured and calculated parameters

Figure 2. Relationship between pressure drop and boil-up rate

Experiment B:

Table 3. Recorded refractive index of dichloromethane at different concentrations

Figure 3: Refractive index vs mole fraction of dichloromethane

The compositions of the mixture were determined using the equation obtained from figure 3. The equation is Y = -0.013x² -0.0315x + 1.4768. Where y represents the refractive index and x represents the molar composition of dichloromethane in the mixture. The value for X was found by substituting the value for the refractive index for a given power input and solve the quadratic equation to determine the molar composition x.

For the first power input (0.65kW)

Overhead RI = 1.4235 so Y = 1.4235   then the equation becomesː

1.4235 = -0.013x² -0.0315x + 1.4768         $\to$

0.04x² + 0.0315x – 0.00533

By solving the quadratic equation the values for x can be obtained.

X₁ = 0.73         X₂ = -0.138

Bottom RI = 1.449 so Y = 1.449   then the equation becomesː

1.449 = -0.04x² -0.0315x + 1.4768         $\to$

0.04x² + 0.0315x – 0.00333

By solving the quadratic equation the values for x can be obtained.

X₁ = 0.41         X₂ = -0.76

This means that the composition of dichloromethane in the overhead product is = 0.41 thus the composition of trichloroethylene is 1-0.41 = 0.59.

For the first power input (0.75kW)

Overhead RI = 1.422 so Y = 1.422   then the equation becomesː

1.4222 = -0.013x² -0.0315x + 1.4768         $\to$

0.013x² + 0.0315x – 0.065

By solving the quadratic equation the values for x can be obtained.

X₁ = 1.23                   X₂ = -0.65

Applying the above procedure for bottom RI, we obtainː

X₁ = 1.6                   X₂ = -0.85

This implies that some errors have affected the experiment because the mole fraction of both components must not be greater than one. The same problem was identified with the next two power increments for both overhead and bottom refractive index (0.85 and 0.95kW).

Experiment C:

The overall column efficiency is defined as the ratio of the total number of theoretical plates and the actual number of plates present in the distillation column as previously stated. The actual number of plates in the column is 8.

Fenske’s method was used to determine the total number of theoretical plates of the distillation column, equation (4) below is the Fenske’s equation used.

N + 1 = $\frac{\mathit{ }\log [\left(\frac{\mathit{XA}}{\mathit{XB}} \right)d*\left(\frac{\mathit{XB}}{\mathit{XA}}\right)b]\mathit{ }}{\log \left(\mathrm{\alpha }\right)\mathrm{av}}$

                                                                           (4)

α_Av = $\surd$

α_d* α_b

α_Av = exp[ $\frac{∆\mathit{Hvap}}{R}(\frac{\mathit{ }1\mathit{ }}{\mathit{Tb}\left(\mathit{tce}\right)}–$ $\frac{\mathit{ }1\mathit{ }}{\mathit{Tb}\left(\mathit{dcm}\right)} \left)\right]$

E = $\frac{\mathrm{Number\; of\; theoretical\; plates}}{\mathrm{Number\; of\; actual\; plates}}*$

100%

T_b(DCM) = 312.9 K

T_b(TCE) = 360.5 K

∆H̅_vap = 27.9J/mol

The actua number of plates is 8.

Power input of  0.65kW

By calculating the value for α and then substituting all the known parameters into equation (4), It was found that the number of theoretical plates for a power of 0.65kW is 3.4.

E = $\frac{3.4}{8}*$

100%                            E = 42.5%

Using McCabe Thiele diagram for distillation of the binary mixture,  as shown in figure 4 below. It was found that the number of theoretical stages is 4. This yields an overall efficiency of 50%.

Figure 4. McCabe Thiele diagram for dichloromethane/trichloroethylene binary mixture

The column efficiency for the power of 0.75,0.85, and 0.95 could not be calculated because the molar compositions of dichloromethane in the mixture was more than one which implies that some technical errors were made while performing the experiment.

Discussion of Results

At the beginning of the experiment, the power was first set to 0.5kW and that power, no pressure drop really occurred. The pressure only began to drop at power>0.6kW. That because the energy that was being generated was not high enough to boil up the solution. When the power was set up at a higher value, the boil-up rate getting higher and the pressure start to drop as the boil- up rate reached its needed rate. However, when the power was set up at 65kW, the pressure starts to drop as the boil- up rate reached its needed rate. As shown in table 1, the boil-up rate (column load) and input power increased as the pressure drop increased. Except for the last input power supply of 0.95 kW was the value of pressure drop decreased. The reason for that might be due to technical errors while performing the experiment as it was quite challenging to record the pressure drop. This situation violates the theory that says that pressure drop is proportional to the column load.

Conclusions

From the results, it is noticeable that there is a constantly increase in pressure drop as the boil-up rate increased. From the results, it is also noticeable that the pressure drop in the column is a function of the input power supplied. The pressure drop increased as the more power was supplied to the column. This would be caused by the energy loses through increased thermal radiation and frictional forces with an increased vapour velocity through the column.

There might be some significant errors that might have affected the results. The values for pressure drop and refractive index are not 100% accurate since the readings were taken using the naked eye. As experiments were only conducted once without repetition for each power input, it is probable that some other results would not be highly accurate.

References:

• Perry’s Chemical Engineering Handbook 7^th ed. 13-12
• Chemical Engineering Vol 1 – Coulson and Richardson
• Chemical Engineering Laboratories Handbook 08-09