KKEK4281

DESIGN PROJECT
Semester 2 Session 2018/2019

Student Name Syed Muzammil Iqbal

Matric No. KEK130703
Assignment Title Mid Semester Report: Design of Vacuum
Distillation Column, DC-203

Group No. 8
Supervisor Dr. Muhamad Fazly Bin Abdul Patah
Course Coordinators Professor Dr. Wan Mohd Ashri Bin Wan Daud
Dr. Fathiah Binti Mohamed Zuki
Submission Date 25 April 2019
Table of Contents

Table of Contents …………………………………………………………………………………. i

List of Figures …………………………………………………………………………………….. ii
List of Tables ……………………………………………………………………………………… iii
List of Abbreviations ……………………………………………………………………………… iv
1.0 Chemical Engineering Design ……………………………………………………………….. 1
1.1 Operating Description ………………………………………………………………………… 1
1.2 Selection of Column …………………………………………………………………………... 2
1.3 Determination of Key Components …………………………………………………………... 3
1.4 Determination of Relative Volatility of Components ………………………………………… 4
1.5 Minimum Number of Stages, Nmin ……………………………………………………………. 4
1.6 Minimum Reflux Ratio, Rmin ………………………………………………………………….. 5
1.7 Theoretical Number of Stages, N ……………………………………………………………... 6
1.8 Overall Column Efficiency …………………………………………………………………… 6
1.9 Feed Tray Location …………………………………………………………………………… 7
2.0 Column Sizing ……………………………………………………………………………….. 7
2.1 Column Height ………………………………………………………………………………... 7
2.2 Column Diameter ……………………………………………………………………………... 8
3.0 Tray Design ………………………………………………………………………………….. 9
3.1 Selection of Tray ……………………………………………………………………………… 9
3.2 Tray Dimensions ……………………………………………………………………………… 11
3.3 Tray Spacing ………………………………………………………………………………….. 11
3.4 Liquid Flow Pattern …………………………………………………………………………… 11
3.5 Weir Design …………………………………………………………………………………..... 12
3.6 Plate Pressure Drop ……………………………………………………………………………. 14
3.7 Flooding Consideration ………………………………………………………………………... 15
3.8 Hole Design ……………………………………………………………………………………. 17
3.9 Tray Layout ……………………………………………………………………………………. 19
References …………………………………………………………………………………………. 20
Appendix A Summary of Chemical Engineering Design of DC-203 ……………………………. A
Appendix B Summary of Tray Design of DC-203 ………………………………………………... C
Appendix C Sketch of Tray Layout of DC-203 ………………………………………………… F

i
List of Figures

Description Page

Figure 1 Mass balance for the DC-203 unit operation 1

Figure 2 Fair’s entrainment correlation plot 5

Figure 3 Liquid flow arrangement as a function of liquid flow rate and column 7
diameter

Figure 4 Weir length correlation in the function of downcomer area, column area 8
and column diameter

Figure 5 Relationship between K2, weir height and minimum height over weir 8

Figure 6 Empirical correlation between orifice coefficient and percent perforated 9

are at different plate thickness to hole diameter ratios

Figure 7 Empirical correlation expressing angle of chord as a function of ratio 10

between weir length and column diameter in graphical plot

Figure 8 Empirical correlation showing the relationship between ratio of net area to 10
perforation area and ratio of hole pitch to hole size in graphical plot

Figure 9 Sketches of the tray layout for distillation column BT-301 A6

ii
List of Tables

Description Page

Table 1 Comparison of type of distillation column. 3

Table 2 Summary of relative volatilities of all the chemical components A1

present in BT-301 system

Table 3 Summary of outcomes from Underwood equations A1

Table 4 Summary of outcomes from Gilliland correlation A1

Table 5 Summary of outcomes from actual number of stages calculations A1

Table 6 Summary of outcomes from Kirkbride approximation A1

Table 7 Summary of the outcomes from column diameter calculations A2

Table 8 Summary of outcomes for tray dimension calculations A3

Table 9 Summary of outcomes for weir design calculations A3

Table 10 Summary of outcomes for pressure drop estimation A3

Table 11 Summary of outcomes from no flooding verification calculation A4

Table 12 Summary of outcomes for hole pitch design calculation A4

iv
List of Abbreviations

Symbol Physical Meaning Units

𝐴a Active area m2
𝜃c Angle of chord °
𝜃d Angle subtended by the edge of plate °
𝐴ca Area of calming zone m2
𝐴c Area of column m2
𝐴u Area of unperforated edge strip m2
𝜇 Average liquid viscosity of the feed stream cP
𝑊 Bottom product flow rate kmol/hr
ton/hr
𝐴m Clearance area under the downcomer m2
𝐸o Column efficiency %
Dc Column diameter m
𝐷 Distillate product flow rate kmol/hr
ton/hr
𝐾j Distribution coefficient of heavy key component
𝐾i Distribution coefficient of ith component
𝐴d Downcomer area m2
ℎb Downcomer backup mm
𝑡r Downcomer residence time s
ℎd Dry plate pressure drop mm
𝑞 Feed liquid fraction
𝐹 Feed rate kmol/hr
ton/hr
𝑢f Flooding velocity m/s
𝐹LV Flow factor
Ψ Fractional entrainment
𝐺 Gilliland coefficient
ℎm Head loss in downcomer mm
ℎap Height of bottom edge of apron above plate mm
𝐻t Height of column m
𝐴h Hole area m2
𝑑h Hole diameter mm
𝑙p Hole pitch mm
𝜌L Liquid density kg/m3
𝜌L,i Liquid density of ith component kg/m3
𝐿 Liquid rate kmol/hr
ton/hr
𝐿T Liquid rate for rectifying section ton/hr

v
𝐿B 𝐿W Liquid rate for stripping section ton/hr
𝜇i Liquid viscosity of ith component cP
ℎb,max Maximum downcomer backup mm
ℎ𝑜𝑤,𝑚𝑎𝑥 Maximum height over weir m
𝑢max Maximum vapour velocity m/s
𝑙s Mean length m
𝑙c Mean length of calming zone m
ℎ𝑜𝑤,𝑚𝑖𝑛 Minimum height over weir m
𝑁min Minimum number of stages
𝑅min Minimum reflux ratio
𝑢min 𝑢h Minimum vapour velocity m/s
𝑀i Molecular weight of ith component kg/mol
𝑥[Link] Mole fraction of heavy key component in bottom product
𝑥HF Mole fraction of heavy key component in feed stream
𝑥D,hk, 𝑥HD Mole fraction of heavy key component in distillate
𝑥I,D Mole fraction of ith component in distillate
𝑥I,F Mole fraction of ith component in feed
𝑥𝐵,𝑙𝑘 𝑥𝐿𝑊 Mole fraction of light key component in bottom product
𝑥LF Mole fraction of light key component in feed stream
𝑥D,lk Mole fraction of light key component in distillate
𝐴n Net area m2
𝑁e Number of stages in rectifying section
𝑁s Number of stages in stripping section
𝑁 Theoretical number of stages
𝐴p Total Area for Perforations m2
𝑃 Operating pressure of distillation column atm
𝑇 Operating pressure of distillation column K
𝑅 Optimum reflux ratio
𝐶o Orifice coefficient
ℎt Pressure drop per plate mm
Pa
𝛼lk Relative volatility of light key component
𝛼ij Relative volatility of ith component relative to heavy key
ℎr Residual head mm
𝐶 Tray spacing m
𝜃 Underwood coefficient
𝜌V Vapour density kg/m3
𝜌V,i Vapour density of ith component kg/m3

vi
𝑦i Vapour mole fraction of ith component
𝑃sat,i Vapour pressure of ith component bar
𝑉 Vapour rate kmol/hr
ton/hr
𝑄 Volumetric flow rate m3/hr
𝑢w Weeping rate m/s
𝐾2 Weir flow coefficient
ℎw Weir height m
𝑙w Weir length m
𝑤c Width of calming zone allowance m
𝑤u Width of unperforated edge strip allowance m

vii
 1.0 Chemical Engineering Design
1.1 Operating Description

Distillate
Temperature 215˚C
Pressure 0.010 bar
Component mass flowrates
(kg/hr)
Palmitic acid 234.30
Stearic acid 324.58
Oleic acid 2000.46
Feed Linoleic acid 270.66

Temperature 242˚C
Pressure 3 bars
Component mass flowrates
(kg/hr)
Palmitic acid 234.73
Stearic acid 334.38
Oleic acid 2023.46
Linoleic acid 272.21
Crude Palm 289.15
Kernel Oil
Bottom
Temperature 258˚C
Pressure 0.013 bar
Component mass flowrates
(kg/hr)
Palmitic acid 0.44
Stearic acid 9.80
Oleic acid 23.00
Linoleic acid 1.55
Crude Palm 289.15
Kernel Oil

Figure 1: Mass balance for the DC-203 unit operation

It is assumed that the process operates at a steady state and the system is ideal. It is assumed that
the inlet temperature of the feed to the column will be below the bubble point of the components.
At this point q>1, considering all the feed to the column is in liquid phase.

The column is operated on a continuous basis. As high temperature can cause severe damage to
the column, the operating pressure of the column will be reduced to 0.013 bar to lower the
boiling point of the components. The operating temperature of the column will be at an average
temperature of 258˚C. This is because the feed has been set to operate at a temperature in
between the boiling points of the components.

The temperature of distillate and bottom product streams will be slightly lower and higher than
bubble point temperature, respectively as total condenser and partial reboiler are being used.

1.2 Selection of Column

There are two main types of columns used for distillation, packed bed column and plate column.
In packed bed columns the gas liquid contact is continuous, however, in plate columns it occurs
stage-wise. In packed columns the liquid flows down the column and over the packed surface
and the vapour flows counter-currently up the column. The adequate operation and performance
of a packed column relies greatly on the maintenance of good liquid and gas distribution
throughout the packed bed. In distillation columns cross-flow plates are the most common type
used. In this type, the liquid flows across the plate and the vapour flows up through the plate.
The liquid is passed from one plate to the next through vertical channels which are known as
downcomers.

The factors discussed below influence the choice between trays & packings. As these are
guidelines for selection of trays or packings for a particular service, it is recommended to
analyse each design case on its own merit for selection. Below is the comparison between these
two types of contacting device:

2
 Table 1: Comparison of type of distillation column.
Packed Bed Column Plate Column

 small-diameter columns (less than 0.6m)  variable liquid and/or vapour loads

 more choices in materials of construction  low liquid rates · large number of stages
for packings especially in corrosive and/or diameter
service (e.g. plastic, ceramic, metal alloys)
 high liquid residence time
 lower pressure drop (important in vacuum
distillation)  dirty service (plate columns are easier to
clean)
 less liquid entrainment · low liquid hold-
up, especially suitable for thermally  presence of thermal or mechanical stress
sensitive material due to large temperature changes which
might lead to cracked packings
 foaming liquids can be handled more
readily (less agitation of liquid by the  exotherms requiring cooling coils inside
vapour) column

Tray column is selected for the design of DC-203 as it is the most common used in large-scale
distillation process as well as its easiness to be cleaned. Also, it can be observed from the mass
flowrates in Figure 1 that the tray column will be more suitable. This is because the flow rates in
this process are large and these will require a large diameter. In cases when the diameter is large
it is more convenient to have plates or trays.

1.3 Determination of Key Components

Before the design stage of a distillation column, key components which are to be separated must
be determined. There are two types of key component; light key (LK) and heavy key (HK). The
light key is described as the component that is desired to be kept out of the bottom product. The
heavy key is described as the component that is desired to be kept out of the top product. Usually
it is relatively easy to determine which the key components are. However, there may be
situations in which close boiling isomers are present so judgement must be used in their
selection. In this case the light and heavy keys are as follows:

3
• Light Key: Oleic acid (A)
• Heavy Key: Trilaurin (B)

1.4 Determination of Relative Volatility of Components

The relative volatility for each component was determined from the ratio between the
distribution coefficients of ith component and heavy component.
𝐾
𝛼𝑖𝑗 = 𝐾𝑖 Eq. (1)
𝑗

Where, 𝛼𝑖𝑗 = relative volatility of ith component with respect to heavy component
𝐾𝑖 = distribution coefficient of ith component
𝐾𝑗 = distribution coefficient of heavy component

Distribution coefficient for each component, however, was determined from the ratio between
the saturated vapor pressure of the component and the total operating pressure of the column.
𝑃𝑠𝑎𝑡,𝑖
𝐾𝑖 = Eq. (2)
𝑃

The saturated vapor pressure for the fatty acids was determined from Antoine equations based on
Antoine coefficients obtained from NIST Chemistry WebBook.

𝐵
𝑃𝑠𝑎𝑡,𝑖 = 𝑒 [𝐴−𝑇+𝐶] Eq. (3)

Where, 𝑃𝑠𝑎𝑡,𝑖 = saturated vapor pressure of ith component (bar)

P = operating pressure (bar)
T = feed temperature (K)
A, B and C = Antoine coefficients for ith component

The relative volatility of all the components are tabulated in Table 2 in Appendix A.

1.5 Minimum Number of Stages, Nmin

For determining minimum number of stages, Fenske equation was used. Fenske (1932) derived a
a shortcut method for determining minimum number of theoretical stages for multicomponent
distillation at total reflux. The derivation assumes that the stages are equilibrium stages. In
finding the minimum number of theoretical stages at total reflux, concentration of LK and HK as
well as relative volatility of LK component is a must-known-information. (Sinnot et. al.,
2005).

𝑥 𝑥
𝑙𝑜𝑔[( 𝐴 ) ( 𝐵 ) ]
𝑥𝐵 𝐷 𝑥𝐴 𝑊
(𝑛 + 1)𝑚 = Eq. (4)
log 𝛼𝐴𝐵

4
Where:
αAB : The relative volatility of light key to heavy key
𝑥𝐴𝐷 : The mole fraction of light key in the distillate
𝑥𝐵𝐷 : The mole fraction of heavy key in the distillate
𝑥𝐵𝑊 : The mole fraction of heavy key in bottom
𝑥𝐴𝑊 : The mole fraction of light key in bottom

The minimum number of stages determined from Fenske equation is 6 stages with the necessary
parameters obtained from mass balance sheet.

1.6 Minimum Reflux Ratio, Rmin

Reflux ratio is defined as the ratio of the liquid (rich in volatile components) that has returned to
the distillation column to the amount of liquid removed as distillate. That means, the higher the
reflux ratio, the higher the liquid flowrate back to the distillation column. Separation between the
components will get better thus few numbers of stages will be required to attain same degree of
separation.
Minimum reflux ratio is an important parameter considered in for the design of distillation
columns. The minimum reflux condition represents the theoretical opposite of total reflux, an
infinite number of ideal separation stages. Minimum reflux ratio is used to determine optimum
reflux ratio of the column. The minimum reflux ratio Rmin is calculated using the Underwood
equation.
𝛼𝑖 · 𝑥𝑖,𝐷
∑𝑖 = 𝑅𝑚𝑖𝑛 + 1 Eq. (5)
𝛼𝑖 − θ

𝛼𝑖 · 𝑥𝑖,𝐹
∑𝑖 =1−𝑞 Eq. (6)
𝛼𝑖 − θ

Where, 𝑥i,D = mole fraction of ith component in distillate

𝑥i,F = mole fraction of ith component in feed
𝜃 = Underwood coefficient
𝑞 = feed liquid fraction

First, underwood coefficient, 𝜃, is determined by performing iteration on equation 6. And then

equation 5 is used to calculate minimum reflux ratio, Rmin. Table 3 in Appendix A summarizes
the outcomes from Underwood equations. From the literature, the optimum reflux ratio is 1.37
times of Rmin for oleochemical processes, due to the similarity in process nature this information
would be employed for the design of DC-203.

5
1.7 Theoretical Number of Stages, N
By using the information obtained from Fenske and Underwood equation, Gilliland correlation is
used to determine theoretical number of stages to achieve desired degree of separation. Gilliland
correlation is given by,
𝑁−𝑁𝑚𝑖𝑛 1+54.4𝐺 𝐺−1
= 1 − exp[(11+117.2𝐺) · ( 𝐺0.5 )] Eq. (7)
𝑁+1

𝑅−𝑅𝑚𝑖𝑛
𝐺= Eq. (8)
𝑅+1

Where, 𝑁 = theoretical number of stages

𝐺 = Gilliland coefficient

The outcomes of Gilliland correlation is shown in Table 4 in Appendix A.

1.8 Overall Column Efficiency, Eo

Overall column efficiency is defined as the ratio of number of theoretical plates and number of
actual plates.
𝑁𝑜. 𝑜𝑓 𝑡ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙 𝑡𝑟𝑎𝑦𝑠
𝐸𝑜 =
𝑁𝑜. 𝑜𝑓 𝑎𝑐𝑡𝑢𝑎𝑙 𝑡𝑟𝑎𝑦𝑠

Eo is applied throughout the whole column, i.e. every tray is assumed to have the same
efficiency. There are several correlations that are used to predict column efficiency, one of the
most widely used correlation is O’Connell expression that can be expressed in the form;
Eo (%) = 51 – 32.5 log (μavg ·αLK ) Eq. (9)
μavg = ∑ 𝑥𝑖 · 𝜇𝑖 Eq. (10)

Where, αLK = relative volatility of light key component

𝑥𝑖 = mole fraction of component in feed
𝜇𝑖 = liquid viscosity of component in feed (Pa.s)

Modified versions of Andrade equation, which introduce one or more constants to improve the
fit to the experimental data have been used to determine viscosities of components. One of the
correlations with higher order terms used is;
𝐵
𝑙𝑛µ𝑖 = 𝐴 + + 𝐶𝑇
𝑇

Where, A, B and C = viscosity coefficients of ith component

6
From the information obtained from column efficiency, actual number of stages can be obtained.
The results are summarized in Table 5 in Appendix A.

1.9 Feed Tray Location

An empirical correlation for estimating feed plate location has been presented by Kirkbride
(1944).
𝑁 𝑥 𝐵 𝑥 2
𝑙𝑜𝑔 𝑁𝑒 = 0.206𝑙𝑜𝑔 [( 𝑥𝐻𝐹 ) 𝐷 (𝑥𝐿𝑊 ) ] Eq. (11)
𝑠 𝐿𝐹 𝐻𝐷

𝑁𝑎 = 𝑁𝑒 + 𝑁𝑠 Eq. (12)

Where, 𝑁𝑒 = number of stages in rectifying section

𝑁𝑠 = number of stages in stripping section
𝑥HF = mole fraction of heavy key component in feed stream
𝑥LF = mole fraction of light key component in feed stream
B = bottom product molar flow rate (kmol/hr)
𝐷 = distillate product molar flow rate (kmol/hr)
𝑥LW = mole fraction of light key component in bottom product stream
𝑥HD = mole fraction of heavy key component in distillate stream

Both equations can be solved simultaneously for Ne and Ns thus determining the feed location.
The results of the correlation are summarized in Table 6 in Appendix A.

2.0 Tower Sizing

2.1 Column Height
The height of a trayed column is calculated by multiplying the number of (actual) stages by the
tray separation. Tray spacing can be determined as a cost optimum but is usually set by
mechanical factors. The most common tray spacing is 24 inches (0.6m). The height of the
distillation column is calculated by the following formula;
𝐶 𝑁
𝐻𝑡 = 0.9 [(𝑁𝑎 − 1) + 10𝑎] Eq. (13)
Where, C = tray spacing (m)

Column height was calculated to be 8.13 m using equation 14.

7
2.2 Column Diameter
The column diameter must be selected so that flooding does not occur, however at the same time
vapor velocities that are high for greater plate are needed. The column diameter is determined
from the flooding correlation for a chosen plate spacing. The superficial vapour/gas velocity
(𝑈nf) at flooding through the net area relates to liquid and vapor densities according to Fair’s
correlation.

Figure 2: Entrainment correlation for sieve plates (Fair, 1961)

𝐶sb is an empirical constant, depending on tray spacing and can be estimated against the flow
parameter (𝐹LG) based on mass flow rate of liquid (𝐿) and vapor (𝑉).
The liquid density is a function of temperature and determined through coefficients. The gas
density on the other hand is determined from aspen.
𝜌𝐿 = ∑𝑖 𝑥𝑖 · 𝜌𝐿,𝑖 Eq. (14)
𝜌𝑣 = ∑𝑖 𝑦𝑖 · 𝜌𝑣,𝑖 Eq. (15)
Where, 𝜌𝐿.𝑖 = liquid density of ith component (kg/m3)
𝜌𝑣.𝑖 = gas density of ith component (kg/m3)
𝑥𝑖 = liquid mole fraction of ith component

8
 𝑦𝑖 = vapor mole fraction of ith component

Typically, the design velocity (𝑈n) through the net area is about 85% of flooding velocity (𝑈nf)
It is a common practice to have uniform tower diameter in all sections of the column even
though the vapor/gas and liquid loadings are expected to be different to minimize the cost of
construction.
The diameter of column is calculated by the following steps;

i. Maximum volumetric flowrate, Q

𝑉𝑚 ∗𝑀𝑤
𝑄= Eq. (16)
3600𝜌
Where, 𝑉𝑚 = vapour rate (kmol/hr)
𝑀𝑤 = Molecular weight of components (kg/kmol)

ii. Net Cross-Sectional Area for Vapor Flow Above the Tray, An
Following the recommendation of estimating column size based on 85% flooding, the
required net tray area is:
1 𝑄
𝐴𝑛 = (0.75) (𝑈 ) Eq. (17)
𝑛
Where, 𝑈𝑛 = design velocity (m/s)

iii. Total Tower Cross-Sectional Area, Ac

Recommended standard tray has a downflow area of 12% of tower area, hence
𝑛 𝐴
𝐴𝑐 = 0.88 Eq. (18)

iv. Column Diameter, Dc

4𝐴𝑡
𝐷𝑐 = √ Eq. (19)
𝜋

The summary of the outcome from these calculations is given in Table 7 in Appendix A.

9
3.0 Tray Design
The tray design of DC-203 is mostly derived from empirical formulas used in fluid mechanics in
graphical means. Tray design comprises of selection of tray type, dimensions of single tray,
liquid flow pattern over the plate, weir design and hole design alongside with verification that
weeping conditions will not take place during normal operation of DC-203.

3.1 Tray Selection

There are three principal cross-flow tray types which are used in plate columns. They are
classified according to the method used to contact the vapour and liquid. They are as follows:

i. Sieve Plate: This type of plate is the simplest type of cross-flow plate. The vapour passes
the holes in the plate and the liquid is retained on the plate due to the vapour flow. In
occasions when flow rates are low liquid weeps through the holes, and this reduces plate
efficiency. Usually the perforations are small holes however, in some cases larger holes
and slots are also made use of.

Figure 3 : Diagram of a Sieve Plate

ii. Bubble-cap Plate: This type of plate is the most traditional and oldest type of cross flow
plate. Various designs have been developed. For most applications the standard cap
design would be specified. In this type of plate, the vapour passes up pipes which are
known as risers. The risers are enclosed by a cap with a jagged edge or slots. Risers
ensure that a level of liquid is maintained on the tray at all vapour flow-rates.

10
 Figure 4: Diagram of a Bubble-cap Plate

iii. Valve Plate: This type of plate is very much like sieve plates however, the only
difference being that they have large diameter holes which are covered by movable flaps.
When the vapour flow increases the movable flaps lift. Valves plates can operate more
efficiently at lower flow rates in comparison to sieve plates. At low flow rates the valves
in the valve plate’s close.

Figure 5: Diagram of a Valve Plate

When selecting the plate type many factors are considered including cost, capacity, operating
range, efficiency and pressure drop. Of the three types sieve plates are the cheapest and are
satisfactory for most applications. The operating costs of sieve plates and the pressure drop is
lower compared to the other types of plates. For these reasons the selected type of plate for the
distillation column is sieve plates.

3.2 Tray Dimensions

The dimensions of tray are derived from the information on chemical engineering design of
DC-203 particularly the column diameter design. With the aid of some governing equations, the
outcomes of the tray dimensions are derived. It is evident that the multiplication factors in the

11
governing equations merely comes from rule of thumb. The tray dimensions along with
governing equations are summarised in Table of Appendix B

3.3 Tray Spacing

Generally, tray spacing ranges from 8 to 36 inches (200 mm to 900 mm). Prime factor in setting
tray spacing is the economic trade-off between column height and column diameter. Most
columns have 600 mm tray spacing. Cryogenic columns have tray spacing of 200-300 mm. Close
spacing is used in small diameter columns, and where head room is restricted; as it will be when
a column is installed in a building. Also, tray spacing help to reduce entrainment. A larger
spacing will be needed between certain plates to accommodate feed and side stream arrangement
and for manways.

3.4 Liquid Flow Pattern

The flow pattern of the liquid on a tray is classified into three types, namely cross flow, reverse
flow and double-pass flow. Cross flow trays are classified according to the number of liquid
passes on the plate. The design shown in Figure 6(a) is a single pass plate. For low liquid flow
rates, reverse flow plates are used. For this type of plate, it is divided by a low central partition
and inlet and outlet downcomers are on the same side of the plate. In double pass plates (Figure
6(c)), the liquid stream is subdivided by using several downcomers, and is used for high liquid
flow rates and large diameter columns. A double pass plate is shown in Figure 6(c). For the
design of DC-203, cross flow trays are used.

Figure 6: Liquid flow patterns on cross flow trays.

12
 Figure 7: Selection of liquid flow arrangement

3.5 Weir Design

i. Weir Height: The height of the weir determines the volume of liquid on the plate and is
an important factor in determining the plate efficiency. A high weir will increase the plate
efficiency but at the expense of a higher plate pressure drop. For columns operating
above atmospheric pressure the weir heights will normally be between 40 mm to 90 mm.
For vacuum operation lower weir heights are used to reduce the pressure drop. For the
design of DC-203, weir height of 6 mm is used as the column is operated at 0.03 bars.

ii. Weir length: With segmental downcomers the length of the weir fixes the area of the
downcomer. The chord length will normally be between 0.6 to 0.85 of the column
diameter. A good initial value to use is 0.77, equivalent to a downcomer area of 12 per
cent. The relationship between weir length and downcomer area is given in Figure 8 .

13
 Figure 8: Relation between downcomer area and weir length

iii. Weeping condition: Weeping occurs at low vapor/gas flow rates. At low vapor flow
rates, liquid start to leak/rain through the perforation (called weeping). The weeping
tendency increases with increasing fractional hole area and liquid flow rates. The
weeping rate can be determined from empirical graphs as suggested by Coulson &
Richardson
Figure 9 shows the relationship between K2, weir height and minimum height over weir.

From Table 9 in Appendix B, both umax and umin are well above the weeping rate, therefore
weeping will not occur during normal operation of DC-203.

14
 Figure 9. Relationship between K2, weir height and minimum height over weir

3.6 Plate Pressure Drop

The pressure drop over the plates is an important design consideration especially for vacuum
distillation. Pressure loss is caused due to two main reasons, vapour flow through the holes (an
orifice loss) and static head of liquid on the plate. The total pressure drop (ht) can be expressed as
follows;
ℎ𝑡 = ℎ𝑑 + (ℎ𝑤𝑐 + ℎ𝑤 ) + ℎ𝑟 Eq. (20)
Where, ℎd = dry plate pressure drop, mm
ℎwc = height of liquid over weir (weir crest), mm
ℎw = weir height, mm
ℎr = residual head, mm

i. Dry plate pressure drop, hd

The pressure drop through the dry plate can be estimated using expressions derived for
flow through orifices.
𝑢 2𝜌
𝑣
ℎ𝑑 = 51 [ 𝐶ℎ ] Eq. (21)
𝑜 𝜌𝑙
Where, 𝐶𝑜 = orifice coefficient as a function of the plate thickness, hole diameter, and
the hole to perforated area ratio.
𝑢ℎ = velocity through the holes (m/s)

𝐶𝑜 can be obtained from Figure 10; which has been adapted by Liebson et al. (1957)

15
 Figure 10: Discharge coefficient, sieve plates (Liebson et al., 1957)

Velocity through holes, 𝑢ℎ is determined by,

𝑄
𝑢ℎ = 𝐴 Eq. (22)
ℎ

Where, Q = vapor flowrate (m3/s)

𝐴ℎ = hole area (m2)

ii. Residual head, hr

Residual head is calculated by simple equation proposed by Hunt et al. (1955);
12.5×103
ℎ𝑟 = Eq. (23)
𝜌𝑙

3.7 Flooding condition

Excessive liquid build-up inside the column leads to column flooding condition. The nature of
flooding depends on the column operating pressure and the liquid to vapor flow ratio. The
column flooding conditions sets the upper limit of vapor velocity for steady operation. There are
three main criteria which determines that the process is operating below flooding point.

16
 i. Criteria 1: Downcomer back-up must be less than half of the tray spacing
The downcomer area and plate spacing must be such that the level of the liquid and froth in
the downcomer is well below the top of the outlet weir on the plate above.
In terms of clear liquid, the downcomer back-up is given by:
ℎ𝑏 = (ℎ𝑤 + ℎ𝑜𝑤 ) + ℎ𝑡 + ℎ𝑑𝑐 Eq. (24)

Where, ℎ𝑏 = downcomer back-up, measured from plate surface (mm)

ℎ𝑑𝑐 = head loss in the downcomer (mm)

Head loss in the downcomer, ℎ𝑑𝑐 is estimated using the equation given by Cicalese et al.
(1947)
𝐿 2
ℎ𝑑𝑐 = 166 [𝜌 𝑤𝑑 ] Eq. (25)
𝐴 𝐿 𝑚
Where, 𝐿𝑤𝑑 = liquid flow rate in downcomer (kg/s)
𝐴𝑚 = downcomer area (m2)

ii. Criteria 2: Downcomer residence time must be greater than 3 s for satisfactory contact
between vapor and liquid.
Enough residence time must be allowed in the downcomer for the entrained vapor to
disengage from the liquid stream. The downcomer residence time is given by:
𝐴𝑑 ℎ𝑏𝑐 𝜌𝑙
𝑡𝑟 = Eq. (26)
𝐿𝑤𝑑
Where, ℎ𝑏𝑐 = clear liquid back-up (m)

iii. Criteria 3: Fractional entrainment should be less than 0.10 to prevent liquid
entrainment within the vapor phase rising upwards.
Fractional entrainment, ψ, can be estimated from the correlation given by Fair (1961), Figure
11.29, as a function of the liquid-vapor factor FLV, with the percentage approach to flooding
as a parameter. The percentage flooding is given by:
𝑢𝑛 (𝑏𝑎𝑠𝑒𝑑 𝑜𝑛 𝑛𝑒𝑡 𝑎𝑟𝑒𝑎)
%𝐹𝑙𝑜𝑜𝑑𝑖𝑛𝑔 = Eq. (27)
𝑢𝑓
Where, 𝑢𝑛 = actual velocity based on net area (m/s)

𝑄
𝑢𝑛 = 𝐴 Eq. (28)
𝑛

By referring to hb, tr and ψ in Table 11, all the three criteria were satisfied and therefore
flooding will not occur in DC-203 operation.

17
 3.8 Hole Design
i. Hole Diameter:
The plate hole diameters (𝑑h) from 3 to 12 mm are commonly used. The bigger sizes are
susceptible to weeping. The holes may be drilled or punched, and the plate is fabricated from
stainless steel and other alloys than carbon steel. For this design, 5 mm hole diameter is used.

ii. Hole Pitch:

Hole pitch is defined as the distance between the center for two adjacent circular holes on the
tray. Hole pitch should not be less than 2 times the hole diameters. From the range the pitch
can be selected to give the number of active holes required for the total hole area. Usually
square and equilateral triangular patterns are used. Of these two the equilateral triangular
pattern is preferred. The total hole area as fraction of the perforated area Ap is expressed in
the following equation.
2
𝐴ℎ 𝑑
= 0.9 [ 𝑙 ℎ ] Eq. (29)
𝐴𝑝 𝑝

Figure 11 and Figure 12 portray the graphical determination of angle of chord and ratio of
hole pitch to hole size. Table 12 in Appendix B summarizes the information obtained from
the calculation performed for this section. The hole pitch determined from the empirical
correlations shows that it is more than twice the hole diameter. Therefore, a stable operation
of DC-203 can be maintained.

Figure 11: Relation between angle subtended by chord, chord height and chord length

18
 Figure 12: Relation between hole area and pitch

3.9 Tray Layout

The tray layout for DC-203 is provided in figure 13 of Appendix C with necessary annotations.
Important parameters such as perforated area and number of holes in tray are also considered.

19
References

20
APPENDIX A SUMMARY OF CHEMICAL ENGINEERING DESIGN OF DC-203

Table 2: Summary of relative volatilities of all the chemical components present in DC-203
system
Chemical Component Relative Volatility, α
Palmitic Acid 5.732883324
Stearic Acid 3.044142597
Oleic Acid 3.569782491
Linoleic Acid 4.685529004
Tricaprylin 46.50067814
Tricaprin 6.028940344
Trilaurin 1
Trimyristin 0.159281418
Tripalmitin 1.79061E-06
Triolein 0.00053738

Table 3: Summary of outcomes for Underwood equations

Variable Equation Value
Underwood Coefficient, Ɵ Eq. (6) 2.28
Minimum Reflux Ratio, Rmin Eq. (5) 1.72
Reflux Ratio, R 𝑅 = 1.37(𝑅min ) 2.36
Liquid flowrate, L 𝐿 23.883 kmol/hr
𝑅=
𝐷

Table 4: Summary of outcomes from Gilliland correlation

Variable Equation Value
Gilliland Coefficient, G Eq. (8) 0.1905
Theoretical Number of Stages, N Eq. (7) 12

Table 5: Summary of outcomes from actual number of stages calculations

Variable Equation Value
Average liquid viscosity, μ Eq. (10) 0.05340607 mPas
Column efficiency, Eo Eq. (9) 74.39 %
Actual number of stages, Na 𝑁 16
𝑁𝑎 =
𝐸𝑜

A
 Table 6: Summary of outcomes from Kirkbride approximation
Variable Equation Value
Number of stages in 11
rectifying section, Ne Eq. (11) & (12)
Number of stages in 5
stripping section, Ns
Feed Tray location - 11th stage

Table 7: Summary of the outcomes from column diameter calculations

Variable Equation Value
Top Bottom
Flow factor, FLV 𝐿 𝜌𝑣 0.015 0.047
𝐹𝐿𝑉 = √
𝑉 𝜌𝑙
Fractional entrainment, ψ Fair’s entrainment 0.27 0.081
(85% flooding velocity) correlation plot
Flooding velocity, unf 𝜌𝑙 − 𝜌𝑣 1.292 m/s 2.488 m/s
𝑢𝑛𝑓 =𝜓√
𝜌𝑣
Volumetric flow rate, Q, Eq. (16) 0.824 m3/s 5.185 m3/s
Area of column, Ac, Eq. (18) 0.85 m2 2.78 m2
Column diameter, Dc, Eq. (19) 1.04 m 1.88 m

B
 APPENDIX B SUMMARY OF TRAY DESIGN OF DC-203

Table 8: Summary of outcomes for tray dimension calculations

Variable Equation Value
Downcomer Area, Ad Ad = 0.12Ac 0.33 m2
Net Area, An An = Ac - Ad 2.45 m2
Active Area, Aa Aa = Ac – 2Ad 2.11 m2
Hole Area, AH AH = 0.1Aa 0.211 m2
Hole Diameter From literature 5.5 mm
Plate Thickness From literature 5 mm
2
Area of one hole, Ah 𝜋(ℎ𝑜𝑙𝑒 𝑑𝑖𝑎𝑚𝑒𝑡𝑒𝑟) 2.376 × 10−5 m2
𝐴ℎ =
4
Number of holes, Nh 𝐴𝐻 8880
𝑁ℎ =
𝐴ℎ

Table 9: Summary of outcomes for weir design calculations

Variable Equation Value
Ratio between weir length Figure 8 0.676
and column diameter, lw/Dc
Weir length, lw - 1.27 m
Maximum height over weir, 2
𝐿𝑤 ⁄3
how,max ℎ𝑜𝑤,𝑚𝑎𝑥 = 750 [ ] 25.394 mm of liquid
𝜌𝑙
𝑙 𝑤
Minimum height over weir, 2⁄
0.7𝐿𝑤 3
how,min ℎ𝑜𝑤,𝑚𝑖𝑛 = 750 [ ] 20.020 mm of liquid
𝜌𝑙 𝑙𝑤
Weir height, `hw From literature 50 mm of liquid
Minimum weir height
(hw + how, min) - 70.020 mm of liquid
K2 Figure 9 30.556
Weeping rate, uw 𝐾2 − 0.9(25.4 − 𝑑ℎ )
𝑢𝑤 =
√𝜌𝑣 9.298 m/s
Maximum vapor velocity, 𝑄
𝑢𝑚𝑎𝑥 =
umax 𝐴𝐻 24.552 m/s
Minimum vapor velocity,
umin 𝑢𝑚𝑖𝑛 = 0.7𝑢𝑚𝑎𝑥 17.186 m/s

C
 Table 10: Summary of outcomes for pressure drop estimation
Variable Equation Value
% Perforated area 𝐴𝐻
× 100
𝐴𝑝 9.1%
Plate thickness to hole 𝑃𝑙𝑎𝑡𝑒 𝑇ℎ𝑖𝑐𝑘𝑛𝑒𝑠𝑠
diameter ratio 𝐻𝑜𝑙𝑒 𝐷𝑖𝑎𝑚𝑒𝑡𝑒𝑟 0.91
Orifice coefficient, Co Figure 10 0.816
Dry plate pressure drop, hd 𝑢ℎ 2 𝜌𝑣
ℎ𝑑 = 51 ( ) ( )
𝐶𝑜 𝜌𝑙 48.887 mm of liquid
Residual head, hr 12500
ℎ𝑟 =
𝜌𝐿 19.294 mm of liquid
Total pressure drop ℎ𝑑 + ℎ𝑟 + ℎ𝑤 + ℎ𝑜𝑤,𝑚𝑎𝑥 143.576 mm of liquid
−3
Pressure drop per plate 9.81 × 10 ℎ𝑡 𝜌𝑙 912.49 Pa

Table 11: Summary of outcomes from no flooding verification calculation

Variable Equation Value
Height of bottom edge of ℎ𝑎𝑝 = ℎ𝑤 − 10 40 mm of liquid
apron above plate, hap
Clearance area under the 𝐴𝑚 = ℎ𝑎𝑝 𝑙𝑤 0.051 m2
downcomer, Am
Head loss in downcomer, hm 𝐿 2 0.027 mm of liquid
ℎ𝑚 = 166 ( )
𝜌𝑙 𝐴𝑚
Downcomer backup, hb ℎ𝑏 = (ℎ𝑤 + ℎ𝑜𝑤 ) + ℎ𝑡 + ℎ𝑚 218.996 mm of liquid
Maximum downcomer ℎ𝑏,𝑚𝑎𝑥 = 0.5𝐶 325 mm of liquid
backup, hb,max
Downcomer residence time, 𝐴𝑑 ℎ𝑏 𝜌𝑙 9.23 s
𝑡𝑟 =
tr 𝐿
Flow parameter, FLV From table 7 0.047
Fractional entrainment, ψ From table 7 0.081

D
 Table 12: Summary of outcomes for hole pitch design calculation
Variable Equation Value
Ratio of weir length to 𝑙𝑤 0.676
column diameter 𝐷𝑐
Angle of chord, ϴc Figure 11 86˚
Angle subtended by edge of 𝛳𝑑 = 180˚ − 𝛳𝑐 94˚
plate, ϴd
Mean length, ls 𝛳𝑑 3m
𝑙𝑠 = 𝜋(𝐷𝑐 − 0.05) ( )
180
Area of unperforated edge 𝐴𝑢 = 0.05𝑙𝑠 0.15 m2
strip, Au
Width of unperforated edge From figure 9 0.05 m
strip allowance, wu
Width of calming zone From Figure 9 0.05 m
allowance, wc
Mean length of calming 𝑙𝑐 = 𝑤𝑢 + 𝑙𝑤 1.32 m
zone, lc
Area of calming zone, Aca 𝐴𝑐𝑎 = 2𝑙𝑤 𝑤𝑐 0.127 m2

Total area for perforations, 𝐴𝑝 = 𝐴𝑎 − (𝐴𝑢 + 𝐴𝑐𝑎 ) 1.833 m2

Ap
Ratio of total hole area to 𝐴ℎ 0.115
perforations area 𝐴𝑝
Ratio of hole pitch to hole Figure 12 2.81
size, Ip/dh
Hole pitch, lp - 15.46 mm

E
 APPENDIX C SKETCH OF TRAY LAYOUT OF DC-203

Cartridge type construction is used. Allows 50 mm unperforated strip round plate edge and
50 mm wide calming zones.
1320 mm

Figure 13: Sketches of the tray layout for distillation column BT-301