“Reading Assignment
Pages 663 through 736 in the McCabe, Smith and Harriot text
Important Points to Consider
(1) Application
Distillation is one of the most important operations in the
chemical processing industries. Many of the mixtures handled by refiners, and
chemical producers can be separated on the basis of boiling point
differences. As long as the relative
volatility of the materials being separated is greater than 1.1 and the
materials are thermally stable, distillation is usually the economical
separation solution. It is one of the
most mature separations with respect to both technology and application. Familiarity and widespread support by
equipment vendors makes distillation a first choice for a number of
applications. It should not be surprising
that distillation is used to perform almost 85 percent of the separation tasks
in the chemical processing industries.
Because it uses energy as the separating agent, no
additional separations are needed. This
usually results in a simple process layout, resulting in a lower capital
investment than would be required for other separations. Distillation has often been faulted for the
amount of energy consumed; however, capital investment must also be considered,
and often proves to be the decisive factor in process selection. Alternative separations are often more
capital intensive when compared with distillation.
Distillation may not be a good choice when the relative
volatility between the components to be separated is less than 1.1. Distillation is ruled out for materials that
are thermally unstable, or that require extremely high or low temperatures to
produce both vapor and liquid phases.
Other separations may be better choices when a dilute high boiling
material is to be recovered. Mixtures
that foam easily can also be poor candidates for separation by distillation.
(2) Design Procedure
The design of a distillation column generally proceeds
through the following stages:
(a) Rates,
feed compositions, product specifications, and operating conditions are
established. These could be modified if
later stages of the design reveal that an infeasible or uneconomical design
results. External material balances are
performed to determine the product stream flow rates and compositions from the
specified feed and separation specifications (recoveries or purities).
(b) The
materials involved and the column pressure determine operating
temperatures. Vacuum can be used to
lower a column temperature where the bottoms temperature is higher than that of
the available heating utility. Elevated
pressure is employed when it is necessary to raise the reflux temperature above
that of the cooling utility. The
material freezing temperatures and critical temperatures determine the feasible
lower and upper temperature bounds for the operation. Liquid column product streams will be at the
bubble point temperature and vapor product streams at the dew point
temperature.
(c) Thermodynamic
models are selected for equilibrium and thermal properties. A poor choice of model can result in
significant errors in the design. For
many systems the relative volatility between the components being separated is
constant and equation 21.4 can be used for the equilibrium relation. It will be evident that this relation is not
linear.
(d) The
required number of equilibrium stages is calculated with a specified value of
the reflux ratio. Studies of
distillation economics have led to the use of reflux ratios that are 15 to 35
percent greater than the minimum reflux ratio.
[Note: when using a software
design package, such as Aspen Plus, you must set the number of stages and
calculate the reflux ratio. This means
that you must adjust the number of stages used until the target reflux ratio is
achieved.] In any case, we have the problem of determining
the minimum reflux ratio. For more
detail see Note A.
(e) Specify
the type of contact internals to be used.
If packing is to be used, either the number of stages must be converted
into a packing height using a Height Equivalent to a Theoretical Plate (stage) value
or the differential contact models developed in Unit 2 must be used.
(f) Perform
the hydraulic design of the trays or packing.
Column diameter is determined. If
undesirable results are generated, the choice in step (e) may need to be reconsidered. See Note D.
(g) Contactor
efficiency is evaluated and the actual number of stages or packing height is
determined. Undesirable results here
could also require the decision in step (e) to be reconsidered. Note that if a rate-based model were used in
step (d), this step would not be needed, as the actual stage performance would
be evaluated directly.
(h) Column
mechanical design is completed.
Condenser, reboiler, and accumulation vessels are designed.
Rating the performance of an existing column requires this
procedure to be followed in the reverse direction. In other words, the column parameters are
specified and the product rates and compositions are to be determined. (Process simulation programs solve these
types of problems more easily than design problems.)
Certain special cases require some additional decisions:
(a)
If the production rate is low, a batch column may be
more economical in operation than a continuous column. Batch columns are widely used in specialty
chemicals production where a variety of products may be produced from the same
facility.
(b)
Steam stripping may be used where the bottoms product
is a dilute aqueous stream. Eliminating
the reboiler might produce cost savings.
(c)
If several streams containing the same species, but
with different compositions, require separation, multiple feed locations can be
provided in a single column.
(d)
If product streams of differing composition are
required, sidestreams can be withdrawn at several locations in the column.
(e)
A partial condenser can be used where a vapor product
is desired or acceptable. Use of a
partial condenser permits the column reflux temperature to be somewhat higher
than that produced with a total condenser at the same pressure.
(f)
A number of column configurations can be produced from the
basic fractionator scheme in order to accommodate special conditions. These might include intermediate condensers
and/or reboilers, sidestream strippers, and simple stripper or rectifier
configurations.
(3) Modeling
Modeling staged distillation columns requires that we move
beyond the simple linear cascade model of the previous unit. The equilibrium relation is rarely linear,
and can be somewhat complex. In fact,
the quality of the results from a distillation model will depend upon the
accuracy of the thermodynamic relations used to generate the equilibrium
relation.
The material balance relations will be linear if the constant molar overflow condition is
assumed. Under constant molar overflow,
the vapor and liquid stream rates in each section of the column are
constant. For this condition to be
realized, the molar heats of vaporization of all chemical species must be
nearly the same, sensible heat effects due to the temperature change through
the column must be small, heat losses must be negligible, and heat of mixing
effects must be insignificant. There are
very few real systems that completely satisfy these requirements; however, many
systems can be modeled as constant molar overflow without producing significant
errors in the results. In other words,
the constant molar overflow assumption is acceptable for making quick
calculations. The widespread use of process simulators for performing
distillation analysis has reduced the need for simple models, but there are
times when a quick, approximate answer is required.
The equilibrium stage approach is the most commonly used
technique for modeling distillation. The
primary distillation algorithms in most commercial process simulation packages
are equilibrium stage based. However,
these same simulation packages are beginning to include rate-based algorithms
that use more fundamental heat and mass transfer rates in the column model, and
are not based upon product equilibrium.
Faster computers have made this approach feasible. The rate-based approach has the advantage of
evaluating separation effectiveness directly, rather than using an efficiency
of some type to correct an equilibrium model.
Note A:
Distillation columns generally have at least two sections
more if there are intermediate products and/or feed streams. [Figure 21.6] This means that the flows in each section
cannot be set independently. For
specified feed and product conditions and flows, only one internal stream flow
can be freely specified. All others will
be determined by the material balances. However,
there is a minimum reflux rate La with respect to the vapor rate Va,
or a maximum boilup vapor rate Vb with respect to the liquid Lb. The minimum rate is determined from a pinch
condition in which two passing streams (vapor and liquid) are in
equilibrium. In many cases this will
occur at the feed stage, but can occur elsewhere in the column if the
equilibrium relation is nonideal.
However, where the relative volatility is constant, the pinch
condition occurs at the feed. The vapor
entering the feed stage will be in equilibrium with the liquid leaving it. These compositions can be used with material
balances written from the feed tray to one end of the column in order to
determine the liquid and vapor flows at the feed tray. Once the minimum flows (reflux liquid or
boilup vapor) are known, they can be increased by an appropriate factor to set
the actual reflux or boilup rates. Specifications
are often made in terms of the external reflux ratio (La/D) or the boilup ratio
(Vb/B).
Once the internal stream flows are determined, the stage
calculation can proceed. Each column
section will have a material balance relation (operating line). The operating lines intersect on a so-called
q-line (page 679). There is one such
line for every feed or side stream. It
will be necessary to determine the intersection points (x and y values) in
order to evaluate the stage operation.
The stage calculation proceeds from one end of the column.
First, the equilibrium relation is used to determine the composition of the
other equilibrium product from the first stage.
The material balance relation is next used to determine the composition
of the other passing stream that is entering the stage. Since this stream is an equilibrium product
from the next stage, this process of using the equilibrium relation followed by
the material balance can be repeated to evaluate all of the stages in the first
section (stripping or rectifying) of the column. In order to determine where the next column
section begins, the stream compositions must be compared to the intersection
points of the operating lines. Once an
intersect point is passed, a new column section begins and a new material
balance (operating line) relation should be used. This total procedure is followed until the
other end of the column is reached the terminal product composition is either
achieved or surpassed. Each time an
equilibrium/material balance calculation is performed a stage is determined
count the total number of times this is done to determine the number of
stages. Feed or intermediate product
stages occur where the calculations switch from one operating relation to
another. [Note: if the column sections are switched each time
an intersection point of the operating relations is encountered, the feeds and
products will be at their optimal locations in the column. If you are rating an existing column, the
number of stages in each section is known and operating relations are switched
whenever the specified number of stages has been evaluated.] By following this procedure, all of the
internal stream compositions will be determined as well as the number of
equilibrium stages.
If a partial condenser and/or a partial reboiler are used,
these devices act as equilibrium separators and will be part of the stage
count. Keep this in mind when
determining the number of contacting trays.
Note B:
Efficiency correction the efficiency concept is used to
determine the actual number of contact stages from the number of ideal
stages. (page 712) The overall efficiency is used to determine
the total number of stages from the total number of equilibrium stages. The Murphree efficiency is applied at each
stage. To do this the procedure in Note
A is modified by adding a step. After
the equilibrium relation is used, the result is corrected by use of the
Murphree efficiency, and then the material balance calculation is made. This process can be observed in the solution
to problem 21.4.
Note C:
If the constant molar overflow assumption is not valid, the
energy balance must also be solved for every stage. This makes the calculation more difficult,
and such problems are best handled with column design software. However, even for simple designs it will be
necessary to solve energy balances around the condenser, reboiler, and any
intermediate exchangers to evaluate the column energy demand.
Note D:
Several approaches can be taken to determine the column
diameter. One of the traditional
approaches is to use the flooding correlation (Figure 21.26). The liquid to vapor ratio and the phase
densities are used to evaluate the abscissa, and the ordinate is found from the
desired tray spacing curve ( 2 ft is common).
The Kv-factor is used in relation 21.68 to determine the allowable vapor
velocity. Dividing the volumetric vapor
flow in the column by the allowable velocity produces a column area. This area should be increased by an
appropriate factor to account for area taken up by downcomers. An alternative approach is to use an F-factor
(21.75) that is determined from a tray efficiency correlation (Figure 21.33 for
sieve trays). In either case, the
evaluation of column hydraulics is not trivial.
Final design evaluations will require specialized analysis software, but
the methods presented here will be adequate for an initial design evaluation.
Learning Objectives
Upon completion of this unit of material you should be able
to:
(1) solve
the material balances to determine product stream flows and compositions
(2) calculate
energy demand for a column
(3) determine
the minimum number of stages and reflux required for a separation
(4) locate column feeds for best separation
(5) determine
the number of equilibrium stages required for a separation
(6) use
overall and plate efficiencies to determine the actual number of contact stages
(7) determine
suitable column diameter based on hydraulic and efficiency considerations
(8) handle
column modifications such as side-streams and partial condensers
(9) evaluate the operation of a simple batch still
Example Problems
Butane Pentane Column
A feedstock containing 32.1 mol % n-butane and 67.9 mol %
n-pentane is supplied as a saturated liquid at the optimum location of a
fractionating column operating at 50 psia.
The feed rate is 100 lb mol/hr, and a distillate product containing 88.5
mol % n-butane is removed at the rate of 30 lb mol/hr. The condenser removes 1.0 million Btu/hr and
produces saturated-liquid product and reflux streams. The latent heat of vaporaization of the feed
is 8500 Btu/lb mol. Assuming constant
molal overflow, determine:
(a)
The reflux ratio a heat balance is written for the
condenser and solved to find the vapor rate in the column overhead. A material balance at the reflux splitter
permits the liquid rate to be calculated, and then the reflux ratio.
(b)
The vapor and liquid rates above and below the feed
the feed is liquid so the vapor rates above and below the feed are considered
to be the same as the vapor rate at the top of the column, liquid rate above is
the same as the reflux rate and the liquid rate
below is the reflux flow plus the feed.
(c)
Number of stages the operating lines are located
using the compositions of the distillate and bottoms streams and the liquid and
vapor rates in each column section. The
operating lines intersect on the q line (vertical for liquid feed). The feed stage occurs at or near the
intersection of the operating lines (optimum location). Starting at the top of the column, calculate
the equilibrium product composition.
Then use the material balance (operating line) to calculate the passing
stream (vapor stream entering the stage).
Repeating this calculation allows the equilibrium stages to be
determined.
(d)
Composition of the bottom product is found from the
overall material balance.
Ethanol_Water Distillation
A fractionating column operating at 1 atm pressure is
supplied with 100 lb mol/hr of a feed that is 25 mol % saturated vapor. The feed is 40 mol % ethanol, and 60 mol %
water, and enters on the optimum stage.
The distillate is saturated liquid containing 80 mol %. The bottom product is saturated liquid
containing 5 mol % ethanol. The reflux
ratio is 1.5 times the minimum rate.
(a)
Minimum reflux ratio find the pinch condition the
operating line of the rectifying (top) section intersects the equilibrium
relation. This occurs near the top of
the column due to the nonideal behavior of the liquid. The passing vapor and liquid at this point
are in equilibrium. The minimum reflux
ratio is found from the operating line relation. Then the actual operating line is located
using the actual reflux ratio.
(b)
Vapor and liquid rates above and below the feed are
found from the distillate and bottoms rates (found from overall material
balances), the assumption of constant flows within each column section, and the
amounts of vapor and liquid added at the feed stage.
(c)
Heat loads on condenser and reboiler found by writing
energy balances around each heat transfer device, using the vapor rates within
each device.
(d)
Equilibrium stages are found from an alternating
solution of the equilibrium relation and the material balance relation. Shown in the diagram as the steps that
occur between the operating lines and the equilibrium relation,
(e)
Minimum number of stages occurs at total reflux (no
product withdrawal), operating lines lie on the 45 degree line. These steps are shown.
Methanol_Water Distillation
A 40 mol % methanol in water feed (50 % liquid) is fed to a
distillation colum at the optimum location.
The distillate is 90 mol %
methanol, and the bottoms product is 10 mol % methanol. The column uses direct steam injection at a
rate 20 % greater than the minimum requirement.
A partial condenser is used.
(a)
Reflux ratio –
Distillate and Bottoms rates are found from the overall material balance
for specified feed rate. The minimum steam
rate is found by locating the operating relation for the stripping section
(dashed line) so that the operating relation intersects the equilibrium
relation on the q line. The slope of
this line provides the maximum liquid to vapor ratio for the stripping section,
and the actual ratio is 20 % lower. The
rectifying operating line is located using the distillate compostion and the
fact that the operating relations intersect at the q line.
(b)
Number of stages if the overall stage efficiency is 50
% – perform the alternating solution of
the equilibrium relation /operating line relation. The partial condenser will account for one of
the equilibrium stages, leaving 6.5
equilibrium stages or 13 actual stages.
(c)
The column diameter is calculated using the vapor rate
and the allowable vapor velocity. The
allowable velocity is determined from the empirical correlation provided in the
textbook. The result is a rather small
column diameter.
Rectifying Column Analysis
A rectifying column contains 16 actual plates (overall
efficiency of 50 %), and is fed a saturated vapor containing 38 mol % A and 62
% B. The saturated liquid distillate is
80 mol % A, and the bottom product contains 30 mol % A. An intermediate liquid product is also
removed. The column produces the maximum
separation with a reflux ratio of 10.9.
(a)
Composition of the side stream the column contains 8
equilibrium stages. The column contains
two cascade sections, and the operating line of the upper section is located
from knowledge of the distillate composition and the slope of the operating
relation. The composition of the passing
streams at the bottom of the column is known, but the operating line slope is
not. However, the two operating
relations must intersect on a vertical q line corresponding to the side
stream. The side stream location is
adjusted until 8 stages can be located in the two column sections. This occurs where the side stream contains 42
% A.
(b)
Liquid/vapor ratios above and below the side stream
these can be determined from the slopes of the respective operating relations.
(c)
Side stream location obtained from the stage
construction, 5th equilibrium stage from the top or 10th
actual plate.
Additional problems that demonstrate different column
configurations:
Benzene Toluene Fractionation two feeds vs. one combined
feed
Rectifying Column w Side Reboiler use of intermediate heat
transfer
Rectifying Column w Intermediate Condenser alternative use
of intermediate heat transfer
Assessment Tasks:
Problem 1. Reconsider the Butane Pentane column
problem. Increase the distillate rate to 50 mol/hr with the same product
purity. Maintain the same reflux ratio. Determine the new condenser heat load, the
new bottoms product composition, the new internal vapor and liquid rates, and
the number of stages required.
Problem 2. Reconsider the Methanol Water
Distillation. If the feed stream
contains 60 mol % methanol, reevaluate reflux ratio, number of stages required,
and the column diameter.
Problem 3. Reconsider the Rectifying column. In this case, no side stream is removed, but
all other specifications remain the same except for the bottoms product
composition. Determine the bottoms
product composition that results for the fixed number of stages.
Pages 663 through 736 in the McCabe, Smith and Harriot text(1) ApplicationDistillation is one of the most important operations in the
chemical processing industries. Many of the mixtures handled by refiners, and
chemical producers can be separated on the basis of boiling point
differences. As long as the relative
volatility of the materials being separated is greater than 1.1 and the
materials are thermally stable, distillation is usually the economical
separation solution. It is one of the
most mature separations with respect to both technology and application. Familiarity and widespread support by
equipment vendors makes distillation a first choice for a number of
applications. It should not be surprising
that distillation is used to perform almost 85 percent of the separation tasks
in the chemical processing industries.Because it uses energy as the separating agent, no
additional separations are needed. This
usually results in a simple process layout, resulting in a lower capital
investment than would be required for other separations. Distillation has often been faulted for the
amount of energy consumed; however, capital investment must also be considered,
and often proves to be the decisive factor in process selection. Alternative separations are often more
capital intensive when compared with distillation.Distillation may not be a good choice when the relative
volatility between the components to be separated is less than 1.1. Distillation is ruled out for materials that
are thermally unstable, or that require extremely high or low temperatures to
produce both vapor and liquid phases.
Other separations may be better choices when a dilute high boiling
material is to be recovered. Mixtures
that foam easily can also be poor candidates for separation by distillation.(2) Design ProcedureThe design of a distillation column generally proceeds
through the following stages:(a) Rates,
feed compositions, product specifications, and operating conditions are
established. These could be modified if
later stages of the design reveal that an infeasible or uneconomical design
results. External material balances are
performed to determine the product stream flow rates and compositions from the
specified feed and separation specifications (recoveries or purities).(b) The
materials involved and the column pressure determine operating
temperatures. Vacuum can be used to
lower a column temperature where the bottoms temperature is higher than that of
the available heating utility. Elevated
pressure is employed when it is necessary to raise the reflux temperature above
that of the cooling utility. The
material freezing temperatures and critical temperatures determine the feasible
lower and upper temperature bounds for the operation. Liquid column product streams will be at the
bubble point temperature and vapor product streams at the dew point
temperature.(c) Thermodynamic
models are selected for equilibrium and thermal properties. A poor choice of model can result in
significant errors in the design. For
many systems the relative volatility between the components being separated is
constant and equation 21.4 can be used for the equilibrium relation. It will be evident that this relation is not
linear.(d) The
required number of equilibrium stages is calculated with a specified value of
the reflux ratio. Studies of
distillation economics have led to the use of reflux ratios that are 15 to 35
percent greater than the minimum reflux ratio.
[Note: when using a software
design package, such as Aspen Plus, you must set the number of stages and
calculate the reflux ratio. This means
that you must adjust the number of stages used until the target reflux ratio is
achieved.] In any case, we have the problem of determining
the minimum reflux ratio. For more
detail see Note A.(e) Specify
the type of contact internals to be used.
If packing is to be used, either the number of stages must be converted
into a packing height using a Height Equivalent to a Theoretical Plate (stage) value
or the differential contact models developed in Unit 2 must be used.(f) Perform
the hydraulic design of the trays or packing.
Column diameter is determined. If
undesirable results are generated, the choice in step (e) may need to be reconsidered. See Note D.(g) Contactor
efficiency is evaluated and the actual number of stages or packing height is
determined. Undesirable results here
could also require the decision in step (e) to be reconsidered. Note that if a rate-based model were used in
step (d), this step would not be needed, as the actual stage performance would
be evaluated directly.(h) Column
mechanical design is completed.
Condenser, reboiler, and accumulation vessels are designed.Rating the performance of an existing column requires this
procedure to be followed in the reverse direction. In other words, the column parameters are
specified and the product rates and compositions are to be determined. (Process simulation programs solve these
types of problems more easily than design problems.)Certain special cases require some additional decisions:(a)
If the production rate is low, a batch column may be
more economical in operation than a continuous column. Batch columns are widely used in specialty
chemicals production where a variety of products may be produced from the same
facility.(b)
Steam stripping may be used where the bottoms product
is a dilute aqueous stream. Eliminating
the reboiler might produce cost savings.(c)
If several streams containing the same species, but
with different compositions, require separation, multiple feed locations can be
provided in a single column.(d)
If product streams of differing composition are
required, sidestreams can be withdrawn at several locations in the column.(e)
A partial condenser can be used where a vapor product
is desired or acceptable. Use of a
partial condenser permits the column reflux temperature to be somewhat higher
than that produced with a total condenser at the same pressure.(f)
A number of column configurations can be produced from the
basic fractionator scheme in order to accommodate special conditions. These might include intermediate condensers
and/or reboilers, sidestream strippers, and simple stripper or rectifier
configurations.(3) ModelingModeling staged distillation columns requires that we move
beyond the simple linear cascade model of the previous unit. The equilibrium relation is rarely linear,
and can be somewhat complex. In fact,
the quality of the results from a distillation model will depend upon the
accuracy of the thermodynamic relations used to generate the equilibrium
relation.The material balance relations will be linear if the constant molar overflow condition is
assumed. Under constant molar overflow,
the vapor and liquid stream rates in each section of the column are
constant. For this condition to be
realized, the molar heats of vaporization of all chemical species must be
nearly the same, sensible heat effects due to the temperature change through
the column must be small, heat losses must be negligible, and heat of mixing
effects must be insignificant. There are
very few real systems that completely satisfy these requirements; however, many
systems can be modeled as constant molar overflow without producing significant
errors in the results. In other words,
the constant molar overflow assumption is acceptable for making quick
calculations. The widespread use of process simulators for performing
distillation analysis has reduced the need for simple models, but there are
times when a quick, approximate answer is required.The equilibrium stage approach is the most commonly used
technique for modeling distillation. The
primary distillation algorithms in most commercial process simulation packages
are equilibrium stage based. However,
these same simulation packages are beginning to include rate-based algorithms
that use more fundamental heat and mass transfer rates in the column model, and
are not based upon product equilibrium.
Faster computers have made this approach feasible. The rate-based approach has the advantage of
evaluating separation effectiveness directly, rather than using an efficiency
of some type to correct an equilibrium model.Note A:Distillation columns generally have at least two sections
more if there are intermediate products and/or feed streams. [Figure 21.6] This means that the flows in each section
cannot be set independently. For
specified feed and product conditions and flows, only one internal stream flow
can be freely specified. All others will
be determined by the material balances. However,
there is a minimum reflux rate La with respect to the vapor rate Va,
or a maximum boilup vapor rate Vb with respect to the liquid Lb. The minimum rate is determined from a pinch
condition in which two passing streams (vapor and liquid) are in
equilibrium. In many cases this will
occur at the feed stage, but can occur elsewhere in the column if the
equilibrium relation is nonideal.
However, where the relative volatility is constant, the pinch
condition occurs at the feed. The vapor
entering the feed stage will be in equilibrium with the liquid leaving it. These compositions can be used with material
balances written from the feed tray to one end of the column in order to
determine the liquid and vapor flows at the feed tray. Once the minimum flows (reflux liquid or
boilup vapor) are known, they can be increased by an appropriate factor to set
the actual reflux or boilup rates. Specifications
are often made in terms of the external reflux ratio (La/D) or the boilup ratio
(Vb/B).Once the internal stream flows are determined, the stage
calculation can proceed. Each column
section will have a material balance relation (operating line). The operating lines intersect on a so-called
q-line (page 679). There is one such
line for every feed or side stream. It
will be necessary to determine the intersection points (x and y values) in
order to evaluate the stage operation.The stage calculation proceeds from one end of the column.
First, the equilibrium relation is used to determine the composition of the
other equilibr”
Recent Comments