This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Distillation is the most commonly encountered separation unit operation in the chemical engineering industry, and is used to separate mixtures based on differences in their volatilities. In this experiment, a batch distillation column unit was operated under total reflux, to investigate its start up, and establishment of equilibrium. A methanol/water mixture was used, and the column contained 8 plates and a reboiler (9 stages overall).
Temperatures at each of the stages were measured, and the apparatus was left until they remained constant, thus equilibrium had been established. The composition of the top product and reboiler were then obtained by measuring their specific gravities, which allowed their volumetric compositions to be determined from a chart. The number of theoretical stages and efficiency of the column were then calculated.
The results showed that the percentage by volume of methanol was greatest in the top product- 47.24% compared with 12.41% in the reboiler. Therefore the mole fraction of methanol was greatest at the top of the column, thus vapour enrichment occurred. This was expected since methanol was the more volatile component.
The theoretical number of stages was determined to be 2, and therefore the column's efficiency was 0.22. This low efficiency may have been due to the heater operating at too high a temperature, and therefore boiling the mixture too quickly, resulting in large amounts of water vapour being released. Therefore, the top product contained considerable amounts of water. The plates may also have lost efficiency due to fouling and wear. The column was poorly insulated and so heat losses would have occurred, reducing the column's efficiency.
There were various errors and assumptions which would have affected the readings and results obtained. It was assumed that methanol and water were the only components present, and that they formed an ideal solution, which is untrue since it is non- ideal. If other components were present, they would have affected the readings.
The specific gravity meter may not have been calibrated accurately, and since the percentage volumes of methanol were read roughly from a chart, they were subject to error. Also, readings from the xy- diagram were subject to reading error, and the diagram assumed constant pressure which was untrue.
The temperatures continued to fluctuate at the end, suggesting that equilibrium may not have been fully achieved. Also, since the top sample was collected over a long time, it is likely that the sample would have contained condensate from before equilibrium was achieved.
Distillation is the most commonly encountered separation unit operation in the chemical engineering industry. It is a mass transfer operation, and is used to separate mixtures based on differences in their volatilities.
Vapour and liquid phases are brought together and eventually, equilibrium is established. This occurs when both have the same chemical potential, hence no longer further mass transfer takes place. The mole fraction of the more volatile component in the vapour phase will be higher than the mole fraction of it in the liquid phase- this is vapour enrichment.
Distillations can be classified as continuous or batch. With continuous, a continuous feed stream is separated to produce two continuous product stream- the top and bottom products. With batch, only a top product is produced- there is no continuous feed stream, and no bottom product is produced.
Batch distillations (this experiment), are often used rather than continuous for convenience, ease of engineering and versatility, and are often used if throughput is small.
Batch and continuous distillations can then be classed as single or multi-stage and multi-stage generally achieves a much better separation. Single stage distillations are commonly used in alcohol production (such as whisky and vodka), and multi-stage distillations are more commonly encountered in industry, particularly in petrochemicals.
In this experiment, a batch distillation with 8 plates and a reboiler (9 stages overall), is used to distil a methanol/water mixture, under total reflux conditions. The start-up and establishment of equilibrium is investigated, and the compositions of the top product and reboiler are obtained. This then allows the number of theoretical stages and column efficiency to be determined.
Since distillations are commonly encountered in industry, an understanding of their design and principles is essential. This experiment reinforces concepts introduced in Process Analysis (2nd year), such as vapour-liquid equilibrium, and also basic concepts introduced in Plant and Process Design (3rd year). More detailed theory is currently being introduced in Separation 2.
To investigate the start-up and establishment of equilibrium in a batch, multi-stage distillation column.
To determine the theoretical number of stages and overall efficiency of the column.
To gain an understanding of mass transfer, distillation theory and the use of xy diagrams.
To improve at analysing experimental procedures, results and errors.
To gain experience in the safe operating of laboratory equipment
To work well as a team and improve report- writing skills.
To gain a greater insight into applications, and the importance of distillation in chemical engineering
Estimation of theoretical minimum number of stages:
With a total reflux, all liquid produced from condensed vapour leaving the column is returned at the top of the column- there is therefore no top product.
The minimum number of theoretical stages can be calculated from the Fenske equation:
xid = mole fraction of the most volatile component (i) in distillate (Methanol is more volatile than water)
xjw = mole fraction of the least volatile component (j) in waste
xiw = mole fraction of the most volatile component in waste
xjd = mole fraction of the least volatile component in distillate
αij = relative volatility of component i relative to component j
(For batch, the waste is the liquid remaining in the reboiler.)
Volatility is defined as:
therefore, with respect to individual components:
Relative volatility is defined as:
Where: Ki = volatility of the most volatile component.
Kj = volatility of the least volatile component.
It is therefore a ratio of the volatilities of the two components and if close to one, the distillation will be difficult as components will have similar boiling points.
Since the relative volatility varies up the length of the column, a geometric mean of these values should be used for the Fenske equation.
The efficiency is defined as:
Experimental Apparatus and Procedure (See Appendix for larger diagrams)
The batch multistage distillation column (operating under total reflux) consisted of 9 stages- 8 plates plus a reboiler. There were thermocouples at each stage, allowing the temperatures to be measured, using a temperature measuring device. A manometer was present to allow the pressure drop across the column to be measured.
It was checked that the liquid level in the reboiler was sufficient to cover the heating element. It was also checked that the manometer contained water. It was ensured that all valves were closed and that the column was operating at total reflux. The electric power was then switched on.
The cooling water supply to the condenser was switched on, and a flow of 2 l/min was established using a flow-meter. The reboiler's heater was switched on, and set to 1.01kW.
Every 5 minutes, the temperature at each stage was recorded using a temperature measuring device. The pressure drop across the column was also measured using a manometer.
After 20 minutes, the reboiler's heater power was reduced to 0.40 kW. It was then increased to 0.52 kW after 70 minutes.
The equipment was left until the temperatures at each stage remained constant-therefore equilibrium had been established.
Samples of the top product and reboiler mixture were collected. They were then placed in ice to cool to 20oC.
After cooling, the specific gravities of the samples were determined using a specific gravity meter.
A chart was used to determine the volumetric composition of methanol in each sample, using the specific gravities measured.
The mole fractions of methanol and water in both phases were calculated for each sample (an xy diagram was used to obtain the vapour fractions from the calculated liquid fractions). This allowed the volatilities and relative volatility to be obtained.
Using these, the minimum theoretical stage number and therefore overall efficiency were determined.
Results and Calculations
Distillate- Top Product
Specific Gravity (kg/m3)
% by Volume Methanol
% by Volume Water
Data used for calculations:
Distillate Sample Calculations: (See Appendix for reboiler sample calculations)
A basis of 1m3 was chosen.
The masses of the components were determined:
The number of moles and mole fractions of each phase were determined:
The vapour mole fractions were obtained from an xy diagram:
The volatility of methanol and water were calculated:
The relative volatility was calculated:
This was repeated for the reboiler sample. (See Appendix)
The geometric mean relative volatility was calculated:
The minimum number of theoretical stages was then calculated: (w and d refer to waste and distillate respectively)
Therefore 2 theoretical stages.
The efficiency was then determined:
Therefore, the theoretical number of stages was 2, thus a reboiler and 1 plate would be required. The efficiency was determined to be 0.22.
Temperature Profiles: Pressure Drop:
The temperature profiles show that temperatures were greatest at the bottom of the tower.
When a large increase in temperature across the tower occurred, a larger pressure drop was measured- this was when the mixture started to boil.
Pressure Drop Sample Calculation: (At 90 minutes)
Manometer reading (h): 90mm ρ= density water (20oC)= 998kg/m3
The pressure drop across the column after 90 minutes was 881 Pa. (Assuming density of air negligible)
The results showed that the percentage by volume of methanol was greatest in the top product- 47.24% compared with 12.41% in the reboiler. Therefore, the mole fractions of methanol were greater in the top product than the reboiler. In the liquid phase, the mole fraction of methanol increased from 0.06 to 0.29, and 0.28 to 0.66 in the vapour phase, thus vapour enrichment occurred. This was expected since methanol was the more volatile component.
The theoretical number of stages was determined to be 2, and therefore the column's efficiency was 0.22, since the column had 9 stages. This low efficiency may have been due to the heater operating at too high a temperature, and therefore boiling the mixture too quickly, resulting in large amounts of water vapour being released. Therefore, the top product contained considerable amounts of water. The plates may also have lost efficiency due to fouling and wear. The column was poorly insulated and so heat losses would have occurred, thus reducing the column's efficiency.
A rapid increase in temperature, and an increased pressure drop was seen after 60 minutes. This would have occurred when the reboiler mixture started to boil, thus releasing hot vapours. As they travelled up the column, passing through the trays, they would transfer heat to the trays (and liquid on the trays). The vapours would then condense, adding to the liquid on the trays. After 60 minutes, the trays would also have reached a sufficient temperature to boil the liquid on the trays, thus releasing more hot vapour.
The temperature decreased up the column which was expected, since the mole fraction of the methanol increased up the column. Being more volatile than water, methanol has a lower boiling point, thus the tower would have operated at a lower temperature at the top.
After the mixture in the reboiler reached its boiling point, a larger pressure drop was measured across the column. After 90 minutes, the temperatures across the column remained constant, thus equilibrium had been reached. At this time, a pressure drop of 881Pa was measured. When boiling, large quantities of vapour would have been produced in the reboiler, and the boiling mixture would have had a larger volume. This would have resulted in increased pressure in the reboiler, and so the pressure at the bottom of the tower would have been greater than at the top, hence the increase in pressure drop.
Throughout the experiment, there were various errors and assumptions which would have affected the readings and results obtained.
The temperature measurement device's readings fluctuated, and thermocouple 3 gave atypical readings. This was likely to be due to poor contact. It is also possible that the device may not have been accurately calibrated, therefore leading to a systematic error. The manometer readings were also a source of reading error, but this would have been very small. The smallest scale division was 1mm, therefore a reading error of ±1/2 mm.
The temperatures and pressure drop were recorded every 5 minutes, but it took a considerable time to carry out the measurements. Therefore, measurements were not taken simultaneously.
The temperatures also continued to fluctuate at the end, thus indicating that equilibrium may not have been fully achieved. The apparatus should have been left for longer to ensure equilibrium had been reached.
The specific gravities measured were also a source of error, as it is possible that the meter wasn't calibrated accurately, leading to a systematic error. The table used to determine methanol composition was also a source of error, as values were taken very roughly from it- no iteration was used. The table was also referenced for 20 oC, but the samples were cooled to 20.5 oC. The values of densities used may also have led to errors, as they too were referenced to 20 oC. It was also assumed that water and methanol were the only components present. If others were present, they would have affected the specific gravities, and therefore the compositions determined, leading to error in the minimum number of plates calculated.
Readings from the xy- diagram were also sources or error, as reading errors may have occurred. Also, the xy- diagram used was based on constant pressure, but the pressure drop measured showed this to be untrue.
These errors and assumptions would have resulted in mole fractions, and therefore the minimum number of plates calculated, being incorrect.
In Fenske's equation, a geometric mean of relative volatities was used, which may not have been a reliable indication of the relative volatility. It was also assumed that methanol and water form an ideal solution, which is untrue since it is non- ideal. This meant that mole fractions may have been inaccurate, and the use of Raoult's law wasn't fully justified.
Since batch distillations are unsteady state, compositions at the top and bottom of the column would have varied with time. Therefore, it is likely that the top sample would have contained condensate obtained before equilibrium was attained. It was also assumed that the column was operating under a total reflux, but it is likely that some liquid would have accumulated in the condenser and piping.
A leak was also observed at the bottom of the column, meaning that liquid was being lost from the system. The column had little insulation, and so heat would have been transferred to the surroundings. This would have decreased the column's efficiency.
Personal Views and Ideas
The experiment was beneficial from an educational point of view, and was relevant and worthwhile. It introduced some important concepts which are currently being introduced in Separations lectures such as equilibrium, mass transfer and the concept of total reflux. It helped clarify Raoult's law and the concept of volatility from Process Analysis. It also covered typical features of distillation columns which were briefly introduced in Plant and Process Design.
The experiment was a good opportunity for me to improve my report writing technique, and gave me an opportunity to evaluate results and errors which are important for an engineer. Distillations are widely used in industry so the experiment was useful for introducing some key concepts behind their operation. It also gave me an opportunity to work in a team. The experiment was well organised and the instructions were clear and accurate. The lab demonstrator was helpful, and explained concepts accurately. The group size was ideal and everyone played an important role.
In the future, the experiment could be improved by using a computer to record all the temperatures. Better results may have been obtained if the heater's power was reduced (but this would take too long).The apparatus could be improved by repairing the leak, and adding additional insulation to minimise heat losses, and improve safety.
It would be interesting to repeat the experiment to observe trends, and to try distilling ethanol and water which form an azeotrope. It would also be interesting to operate the column continuously and with varying refluxes.
On the whole, the experiment was relevant to the course and has improved my understanding and appreciation of the key concepts behind distillation.
The results showed that the percentage by volume of methanol was greatest in the top product- 47.24% compared with 12.41% in the reboiler. Therefore, the mole fractions of methanol were greater in the top product than the reboiler. In the liquid phase, the mole fraction of methanol increased from 0.06 to 0.29, and 0.28 to 0.66 in the vapour phase, thus vapour enrichment occurred. The theoretical number of stages was determined to be 2, and therefore the column's efficiency was 0.22, since the column had 9 stages. This low efficiency was likely to be due to rapid boiling.
There were various errors and assumptions which would have affected the readings and results obtained, such as reading errors from charts, and the assumption that equilibrium had been attained.
A greater understanding of distillation columns, and the theory, calculations, errors and assumptions associated with them was achieved, and thus all objectives set were accomplished.