Ind. Eng. Chem. Res.

2008, 47, 51-65 51

PROCESS DESIGN AND CONTROL

Process Design Approach for Reactive Distillation Based on Economics, Exergy,

and Responsiveness Optimization
Cristhian P. Almeida-Rivera*
Process Science Department, UnileVer Food and Health Research Institute, OliVier Van Noortlaan 120,
3130 AC Vlaardingen, The Netherlands

Johan Grievink
Process Systems Engineering Group, Delft UniVersity of Technology, Julianalaan 136,
2628 BL Delft, The Netherlands

The interactions between economic performance, thermodynamic efficiency, and responsiveness in reactive
distillation process design are explored in this contribution. This motivation is derived from taking a sustainable
life span perspective, where economics and avoidance of potential losses of resources (i.e., mass, energy, and
exergy) in process operation over the process life span are taken into consideration. The approach to reactive
distillation column design involves defining a generic lumped reactive distillation volume element and the
development of rigorous dynamic models for the behavior of the element. As an extension of conventional
existing models, this model, which is used as a standard building block, includes entropy and exergy
computations. Three objective functions are formulated, which account for the process performance regarding
economy, exergy loss, and responsiveness. A fundamental understanding of the strengths and shortcomings
of this approach is developed by addressing various case studies: (i) steady-state simulation of a classical
MTBE design based on economics only, useful as a reference case; the multiobjective optimization of a
MTBE reactive distillation column with respect to (ii) economics and thermodynamic efficiency; and (iii)
economics, thermodynamic efficiency, and responsiveness. Structural differences and dynamic responses are
identified between the optimized and classical designs to stress the importance of considering exergy and
responsiveness criteria. Thus, the classical (economics-driven) design is compared to a green (economics-
and exergy-driven bioptimized) design. The green design produces less entropy than the classical one, while
being marginally less attractive from an economic standpoint. The bioptimized design has an improved closed-
loop performance compared to the classical design, keeping the same control structure and settings. It is
concluded that incorporating both economic- and exergy-related objectives in the unit design results in a
process with better closed-loop performance and reduced exergy loss.

Introduction transport phenomena can, however, lead to undesired, system-

Advances in computing and information technology allow
chemical engineers to solve complex design problems arising
from a need to improve sustainability of the biosphere and
human society. In such a context, the performance of a chemical
plant needs to be optimized over its manufacturing life span,
while accounting for the use of multiple resources in the design
and manufacturing stages. Since such resources are of a different
nature (e.g., capital, raw materials, and labor) with different
degrees of depletion and rates of replenishment, the performance
of a chemical plant is characterized by multiple objectives with
tradeoffs.1 An integrated unit operation like reactive distillation
(RD) can offer significant benefits for resource utilization, when
it is accompanied with the avoidance of chemical equilibrium
limitations, enhancement of conversion, selectivity, and yield,
and the reduction of both operational and capital costs. The
nonlinear coupling of chemical reaction, phase equilibria, and
* To whom correspondence should be addressed. Tel: +31 10 460
6504. Fax: +31 10 460 5025. E-mail: Cristhian.Almeida-Rivera@
Unilever.com.
10.1021/ie061521j CCC: $40.75 © 2008 American Chemical Society
Published on Web 11/30/2007
dependent features, such as multiple steady states2-17 and
reactive azeotropes.18-25
The results of a case study on the interactions between
economics, exergy efficiency, and process responsiveness in the
conceptual design of RD processes are discussed in this
contribution. In the wider context of conceptual process design,
sustainability issues like feedstock selection, (re)use of catalyst,
and solvents are of key importance. In our perspective of life
span, we do not cover those aspects as we focus exclusively on
the minimization of loss of resources in the operational phase.
Currently, conceptual design in chemical engineering is
predominantly driven by (steady-state) economic considerations,
postponing to latter engineering stages any issue related to
exergy analysis, responsiveness, and control performance. The
importance of incorporating exergy efficiency in the early stages
of the design cycle is stressed by the increasing depletion of
nonrenewable energy resources and the relatively low second
law efficiencies (5-20%) for the majority of chemical processes.
In spite of these facts, exergy analysis is in most of the cases
performed as a follow-up task to process synthesis.26-29 In
52 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008

addition to economic and exergy considerations,30 the perfor-

mance optimization of a chemical plant should include control-
lability and process responsiveness. These aspects are important
to maintain product quality and to avoid excessive use of utilities
for control actions and reprocessing of off-spec material.
The focus of this contribution is the importance of including
interactions between economics, exergy efficiency, and process
responsiveness in RD conceptual design. Aiming at these goals, Figure 1. Schematic representation of the GLRDVE. Feed and side draw-
a generic compartment model is defined, which accounts for off streams and heat exchange are allowed.
mass, energy, and momentum balances and includes entropy
and exergy computations. This behavioral model is integrated is triggered by the presence of an appropriate catalyst. Each
to create a RD column structure. The design is cast into a GLRDVE has its own complexities due to the many interacting
nonlinear optimization problem, where degrees of freedom in phenomena. However, this does not account for the interactions
the design model are fixed by optimization of multiple objectives between the stages in a real multistage column. The latter class
and applying constraints, reflecting heuristics and engineering of interactions could give rise to nonlinear dynamic behavior
knowledge. At this column level, three performance criteria are and poor controllability if an unfortunate combination of reflux
defined for the optimization of the unit design. These embrace ratio and number of (non)reactive trays is chosen. Issues related
the unit performance regarding economics, thermodynamic to connecting compartments and superstructure description are
efficiency, and process responsiveness. The economic perfor- addressed in Integration of Volume Elements to a Column
mance index is given by the total annualized cost, whereas the Structure, Definition of Optimization Criteria, and Optimization
thermodynamic efficiency index accounts for the process Approach.
irreversibilities and is expressed in terms of the entropy Mass/Energy/Entropy/Momentum Balances. Component,
production rate by chemical reaction, heat, and mass transfer. energy, and entropy balances are developed in the GLRDVE.
The responsiveness index is similar to the response time defined The entropy and, in particular, the entropy production rate are
by Luyben et al.31 The three objective functions are solved of paramount importance to determine exergy losses and the
simultaneously within a single multiobjective optimization responsiveness of the process. The balances lead to the following
problem. At this level of understanding, it is believed that a set of differential algebraic equations,
multiobjective optimization gives more insight into trade-offs
i ×V )
∂(F(R) (R) j)nst
between objectives than combining the three objectives into a
single monetary objective function using ad hoc weighting
∂t
) ∑
j)1
Fj,(R)
i - Ψi × aΨ + δ(R) × ri × VLRX
factors. Comparing the design based on the 3-fold performance
criteria with a reference economic-driven design allows us to i ∈ Znc (1)
identify structural differences. The response of the optimized j)nst i)nc
∂U(R)
design in the presence of a deterministic disturbance scenario
is validated using dynamic simulation. ∂t
) ∑ ∑
j)1 i)1
Fj,(R)
i × hj,(R)
i + Ψe × aΨ +.... +
The thrust of this contribution is the optimization-based design j)nst i)nc
approach. Multiple objective functions are considered and
embedded in a large-scale optimization problem.
∑ ∑
j)1 i)1
Fj,(R)
i × gj,(R)
i + δ(R) × Q(R) (2)

Generic Lumped Reactive Distillation Volume Element where Q(R) is assumed to only occur in the liquid phase (i.e.,
Q(R) ) QL).
The starting point for a fundamental understanding of the The Kronecker delta function defined as
strengths and shortcomings of the proposed multiobjective
design approach is the definition of a generic lumped RD
volume element GLRDVE. This element (or compartment) is
chosen as a building block in the design approach and selected
δ(R) :) { 0 (if (R) * L)
1 (if (R) ) L)
(3)

to represent the governing phenomena occurring in RD process- The energy flux associated with the interphase mass and heat
ing. This behavioral model accounts for the following: (i) mass, transfer Ψe is composed of a measurable heat flux and the partial
energy, and momentum balances, together with entropy and molar enthalpies brought by the component fluxes. Namely,
exergy computations; (ii) thermodynamic equations of state for i)nc

∑
the specific heat and specific density and for the chemical
potential per species; (iii) thermodynamic phase equilibrium Ψe ) - Ψi × h(R)
i +q
(R′-R)
(4)
i)1
conditions at the interphase; (iv) the constitutive rate equations
for chemical reactions; (v) the rate equations for interphase The entropy balance in the transient state at a macroscopic
transport of species mass and energy; (vi) the rate equation for level is given by the net amount of entropy transferred into the
heat transfer with an external solid surface; (vii) the hydrody- volume element per unit time and the amount of entropy used/
namics-based rate equations for convective mass and energy produced in the control volume per unit time. Thus, the entropy
flows with the external world and the pressure drop; and (viii) balance for a given phase (R) can be stated as,
the equations for the interphase contact surfaces and for the
void fraction of the vapor phase in the liquid phase. j)nst i)nc
∂S(R)
The following attributes are introduced for the system
proposed in Figure 1: (i) two internally (well-mixed) homo- ∂t
) ∑ ∑ Fj,(R)
j)1 i)1
i × Sj,(R)
i + Σ(R)
V (5)

geneous fluid phases; and (ii) chemical reactions in the liquid

phase only. Thus, mass transfer between phases and heat where Σ(R)
V includes exclusively interfacial diffusional and
exchange are allowed in the compartment, and chemical reaction chemical reaction contributions.
 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 53

Alternatively, the entropy balance can be expressed in terms Table 1. Expressions To Estimate Driving Forces in the GLRDVE
of the entropy density s(R) and the fluxes of entropy density,32,33 of Figure 1a
parameter driving force
∂s(R) Xm,i ) -(1/T)∆µi,T ) R × ln(yin
) -∇J(R) i × P /yi × P)
s + σV
(R) mass transfer in
(6)
∂t interphase energy transfer Xe ) ∆(1/TV) ) 1/2((1/Tout,V) - (1/Tin,V))
chemical reaction Xrx,j ) (ALj /T)
For stream j, the flux of entropy J js is related to the total flux external heat exchange XHX ) (1/T) - (2/(Tu,in + Tu,out))
of energy J je and mass fluxes J m,i
j
according to the following a i ∈ Znc; j ∈ Znrx .
expression, 32

For component i the driving force Xm,i is related to the

1
i)nc µj,(R) chemical potential gradient between the incoming and outgoing
∑
i
Jj,(R)
s ) × Jj,(R)
e - × Jm,i
j,(R)
j ∈ Znst (7) vapor streams.37,38 The integrated flux associated with Xm,i is
T i)1 T calculated by mass balance over the vapor phase,28,37
The fluxes of mass and energy required in expression 7 can j)nst
be obtained from the equation of continuity per volume unit
and energy balance, respectively, for an open unsteady-state
Ji × aΨ ) ∑
j)1
total × yi
Fj,V j
i ∈ Znc (11)
multicomponent system.34 The entropy production rate at
macroscopic level Σ(R)V (eq 5) can be obtained after integration
This integrated flux expression is consistent with the com-
of the local entropy production rate density σ(R) V over the
ponent balance given by eq 1 if Ψi equals Jm,i. Moreover, the
volume element in phase (R). For the sake of simplification, total streamflow is related to the component flow according to,
the following assumptions are adopted: (i) entropy contributions
total × yi ) Fi
Fj,V i ∈ Z nc
j j,V
of all involving phases in the GLRDVE are combined in a single (12)
V ) f (ΣV )); (ii) entropy is produced due to mass
variable (Σ(*) (R)

and heat diffusion in the liquid-vapor interphase, chemical
reaction in the bulk of the liquid, and heat transfer from external
heat sources/sinks and the surrounding environment to the liquid
phase only; and (iii) the effects of electric field, external forces,
and diffusive momentum fluxes are neglected. The following
The driving force Xe is estimated as suggested by Koeijer
and Rivero.28 The measurable heat flux through the interphase
associated with Xq is derived from the energy balance in the
vapor side,37
j)nst

∑
expression for the integral entropy production term can be then
derived,32,35,36 Jq × aΨ ) total × hi
Fj,V j,V
(13)
j)1
i)nrx
1
∫ ∫ ∫V σ (*)
dV ) Σ(*)
V )
L
∑ ALi × Li × VLRX + JLq × In this expression, the enthalpy reference state is calculated at
the average temperature at which components are transferred
T i)1
i)nc through the interphase. The expression suggested by Koeijer et
a Ψ × Xq + ∑
i)1
Jm,i × aΨ × hi × Xe + al.39 and Koeijer37 has been used to estimate the XHX average
driving force.
i)nc In the development of the GLRDVE, a reduced form of the
(- ∑
i)1
Jm,i × aΨ × Xm,i) + QL × XHX (8) momentum balance is considered. We assume a pseudo steady
state and negligible viscous contribution. Moreover, the gradient
of pressure is assumed to be much larger than the rate of
The entropy contribution due to chemical reaction contains momentum gain by convention of the fluid. These simplifica-
the affinity of chemical reaction and its degree of advancement, tions applied to the momentum balance expression given
which are defined respectively as, elsewhere,34 result in the expression,
i)nc
∇P(R) ) F(R) ‚g(R)
∑
(14)
ALi ) - νLi,j × µLi j ∈ Znrx (9)
i)1 Availability and Availability Function. Two additional
j)nrx variables of key importance for this thermodynamic-based
rLi ) ∑
j)1
νLi,j × Lj i ∈ Z nc (10) approach are the availability A and the availability function B.
In this context, availability is regarded as a measure of the
maximum amount of useful energy that can be extracted when
To estimate the driving forces and fluxes occurring in the the system is brought to equilibrium with the dead-state
volume element, additional assumptions are introduced: (i) the surroundings.35 Thus, the availability of the system in phase
GLRDVE is composed of incoming and outgoing streams of (R) and state (T1,P1) is given by the expression,
liquid and vapor phases; (ii) the vapor phase behaves ideally;
(iii) the flux of mass and heat through the interphase is positive A(R) ) (h(R) - T0 × S(R))T1,P1 - (h(R) - T0 × S(R))T0,P0 (15)
from the liquid side to the vapor side; (iv) dynamic changes in
gas phase hold-ups of species mass and energy are ignored;
where the dead-state conditions are those at which T0 ) 298.15
and (v) Gibbs free energies at liquid and vapor sides are equal.
K and P0 ) 1 atm.
The underlying reasoning behind the inclusion of these
The availability function B at state given by (T,P) is computed
assumptions is a more transparent derivation of the driving
according to the following equality,
forces and flux expressions, but without compromising the
validity of the approach. The expressions used to estimate the
averaged driving forces are summarized in Table 1.
(R)
BT,P ) (h(R) - T0 × S(R))T,P (16)
54 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008

Contrary to the availability, the availability function of the investment and operational costs of the process,
system at a given state (T1,P1) can be referenced to any other
state (T2,P2), j)nsti)nc

∆B (R)
) (h (R)
- T0 × S )T1,P1 - (h
(R) (R)
- T0 × S )T2,P2 (17)
(R)
fecon ≡ TAC ) ∑ ∑
j)1 i)1
Fj,(R)
i ci + Fcuccu + Fhuchu +

cunit+catalyst R ∈ Znp (19)

According to their definitions, A and B increase with
increasing enthalpy and decrease with increasing entropy. The exergy efficiency fexergy is accounting for the process
Moreover, the numerical values of their variations coincide (i.e., irreversibilities and is expressed in terms of efficiency of the
∆B(R) ) ∆A(R)). second law of thermodynamics, ηII,
Based on the combination of energy and entropy balances
and according to the definition of availability an exergy balance Wideal
can be obtained for a given phase (R) in the absence of external fexergy ≡ ηII ) (20)
forces (see Appendix A in Luyben et al.31), Wideal - T0ΣV

∂B(R) × M(R)
j)nst
where Wideal (Wideal < 0) is defined as by Seider et al.35 and
) ∑ Fj,(R)
total ×B j,(R)
+ fixed for a given process.37

( )
∂t j)1 From a sustainability perspective, the desired goal is the
T0 enhancement of ηII. According to eq 20 the maximization of
1- Q(R) - T0 × Σ(R)
V (18) fexergy is equivalent to the minimization of ΣV.37,40 In this
TQ contribution, we adopt this second approach and compute T0ΣV
for a fixed T0 value.
Design Decision Variables. According to a degree of
The responsiveness index fresponse is approximated to the
freedom analysis performed in the GLRDVE the following
settling time defined by Luyben et al.,31
design variables are to be specified for a given operating
conditions and feed specifications:
∆B(x)
(a) reaction volume V(R) fresponse ≡ τ ) max (21)

{
RX
x T0∆ΣV(x)

V(R)
RX )
V(R) (for liquid phase homogeneously
catalyzed reaction(s)
-1 for heterogeneously catalyzed
mcat × Fm,cat ( reaction(s)
(b) external and heating/cooling flow rate Q(R), and
(c) the interfacial area for heat and mass transfer aΨ.
Connecting streams and associated flow rates Fj,(R) are
i
assumed to be given. Operational variables such as reflux ratio
and pressure are of key relevance in RD and should be chosen
with thoroughness. In our approach, these variables are specified
at a higher decision making layer (cf. Volume Elements to a
Column Structure, Definition of Optimization Criteria, and
Optimization Approach), aiming at operating windows where
steady-state multiplicity and reactive azeotropy phenomena are
not likely to occur.
The perspective adopted for selecting the design decision
variables is task-oriented rather than the conventionally used
equipment-oriented design approach. First, the task-based
concept would allow us to derive the phase volume V(R) from
the reaction time (i.e., residence time in element) to meet a
specified degree of advancement of reaction; second, the amount
of heat to be added or removed allows one to compute the heat
exchange surface and coolant flows; and third, the interfacial
areas for species mass and heat transfer may be translated into
the mechanical design of a tray and its prevailing hydrodynamic
regime such that the required areas can be met within constraints.
The main benefit of this functional, task-oriented perspective
involves the process design unconstrained by equipment ge-
ometry. Thus, task-oriented design prevails against equipment-
oriented design.
Economics/Exergy Efficiency and Responsiveness Objec-
tive Functions. A set of three performance criteria indicators
is defined for the GLRDVE: the economic performance index
fecon, the sustainability performance index fexergy, and the
responsiveness performance index fresponse.
The economic performance indicator fecon represents the total
annualized cost TAC of the GLRDVE, including capital
) At steady-state conditions B(x) is at its minimum value and
) increases when the system is perturbed by a disturbance. Since
a similar situation holds for ΣV(x), the response time τ measures
how fast a small variation ∆B(x) would dissipate due to an
increased ∆ΣV(x).31
In the previous definition of fresponse, Luyben et al.31 assumed
that both the availability and entropy production rate are
exclusively functions of the vector of state variables x. Although
this assumption is fundamentally correct, it somehow disguises
the way of estimating the effect of a small disturbance or a
change in an input on the system. Therefore, we propose a subtle
modification of the arguments of B and ΣV, by which a second
argument is included, representing the systemic view on a
dynamic system. Thus, the availability and entropy production
rate are regarded as functions of an internal state vector x and
a vector of disturbances or manipulated external excitations u,
exergy: B(x,u) (22a)
entropy production rate: ΣV(x,u) (22b)
Combining the definitions of eqs 22 with eq 21 leads to a
more generalized expression for fresponse,
∆B(x,u)
fresponse ≡ τ ) (23)
T0∆ΣV(x,u)
The previous expression could be further elaborated by
defining the values of state variables x and disturbances u at
each state. Hence,
∆B(x,u) ) B(x,u) - B(x0,u0) (24)
∆ΣV(x,u) ) ΣV(x,u) - ΣV(x0,u0) (25)
with

u ) u0 + ∆u (26)
x ) x(u) (27)
ΣV(x,u) g ΣV(x0,u0) (28)
An additional feature of Luyben’s original definition of
fresponse, not shown in expression 23, is related to the maximiza-
tion for the time constant τ over all possible steady states
surrounding the reference steady state. In this way, one looks
for the worst case over all possible small-scale disturbances and
input change scenarios. This optimization-driven approach
justifies Luyben’s approach of including a single argument for
∆B and ∆ΣV.
In our approach, such an maximization approach is not
considered due to its inherent high computational effort.
However, a rather formal definition of fresponse is adopted (eq
23) together with a single input excitation. By not searching all
possible state trajectories, the generality of the results might be
questionable and the external excitation is a fairly conservative
one. However, engineering judgment can be exercised in
selecting the disturbance variable u and the magnitude of the
disturbance ∆u.
Integration of Volume Elements to a Column Structure,
Definition of Optimization Criteria, and Optimization
Approach
Creating a Column Structure from the Generic Volume
Element. The integration of compartments to create a column
structure introduces additional degrees of freedom, which should
be conveniently specified (e.g., reflux ratio). In the more general
context of process synthesis, the issue of coupling compartments
to a process unit has been frequently addressed41-43 and is not,
therefore, explicitly considered in this research. Briefly speaking,
the compartment integration is carried out by introducing a
superstructure and finding the optimal configuration using hybrid
optimization techniques over the superstructure. The description
of the superstructure and the rules for connecting compartments
with each other and with the outer world are embedded in the
structure model. Extensive information on structure models can
be found in the research work of, for example, Ismail et al.41
and Papalexandri and Pistikopoulos.42
The rules of connectivity used in the RD superstructure state
the following: (i) the outgoing liquid at a given compartment
enters the compartment located below; (ii) the outgoing gas at
a given compartment enters the compartment located above;
(iii) the liquid stream leaving the condenser is partially recycled
back to the top compartment and withdrawn from the unit as
top product; (iv) the liquid leaving the bottom compartment
enters the reboiler; (v) the streams leaving the reboiler are a
vapor stream (entering the bottom compartment) and a liquid
stream (withdrawn from the unit as a bottom product); and (vi)
side feed streams and side withdrawal streams may occur.
The connectivity rules, however, impose some configuration
limitations: (i) heat integration over volume elements is not
considered; (ii) side reactors are not present; and (iii) no parallel
structures are considered.
During the development of the RD unit, the following set of
simplifications are made: (i) the liquid and vapor streams
leaving a tray are in thermodynamic equilibrium; (ii) both liquid
and vapor hold-ups are fully considered; (iii) the liquid and
vapor on each tray are perfectly well mixed (i.e., no radial
gradient in temperature and composition and the liquid on the
tray has a concentration equal to that of the liquid leaving the
tray); (iv) the reactions are taking place in the liquid phase and
Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 55
enhanced by a solid catalyst; (v) the reactions occur only on
the trays containing catalyst; (vi) vapor and liquid entrainment
is neglected; (vii) the column is equipped with a condenser,
reflux drum, and reboiler; (viii) the location(s) of the feed
stream(s) and product stream(s) are known a priori; (ix) the
liquid hold-up on a tray is related to the liquid outflow according
to Francis weir formula; (x) the flow of vapor leaving a tray is
determined by the pressure drop between two consecutive trays;
(xi) the tray pressure drop includes contributions of dry pressure
drop, liquid height, and residual head; (xii) the physical
properties of the involved streams are computed as a function
of temperature or phase composition; (xiii) the heat of reaction
is omitted in the energy balance, as the heat of formation at
standard conditions (1.033 × 105 Pa/298 K) is used as reference
in the enthalpy computations; (xiv) the reboiler, condenser, and
reflux drum are nonreactive units; (xv) the tray sizing is
estimated by rules-of-thumb criteria;44 and (xvi) the column
diameter is larger than the minimum flooding value.
The compartments representing the behavior of the condenser
and reboiler are, in principle, identical to the GLRDVE, provided
the following characteristics are present: (i) no chemical reaction
takes place due to the absence of catalyst; (ii) heat exchange
takes place in both phases; (iii) the condenser is a total
condenser; and (iv) the reboiler is a partial reboiler.
In addition to the specification of the reflux ratio, the
compartment integration to the superstructure involves the
following features of the structure model: (i) all compartments
can have a different catalyst load; and (ii) the column sections
can have different diameters.
Development of Three Criteria for Optimization of
Column Design. Stepping out from the GLRDVE case to the
integrated RD unit is intuitive for fecon and fexergy. The variables
∆B(x,u) and ∆ΣV(x,u), on the other hand, need to be globally
evaluated over all the structure for the responsiveness criterion
fresponse. The resulting performance criteria are given by the
following set of expressions for the kth compartment,
k)nt j)nst i)nc
fecon ) ∑ ∑ ∑
k)1 j)1 i)1
i ci + Fcuccu + Fhuchu + cunit+catalyst]
[Fj,k k k k
(29a)
total
(Wideal)
fexergy ) (29b)
(Wideal)total - T0(ΣV)total
(∆B(x,u))total
fresponse ) (29c)
T0(∆Σ V(x,u))total
The validity of an additivity property (i.e., the objective
functions per compartment can be added up to a corresponding
objective function over the entire structure) is implicitly assumed
in these definitions.
An indication of the implications of process responsiveness
for design is revealed by the definition of fresponse. According to
eq 30, a short responsiveness time is desired and can be attained
by a large change in the entropy production rate for a given
change in stored exergy. Using an ideal heat exchanger as a
case study, for instance, Almeida-Rivera45 and Meeuse46
demonstrated that better process responsiveness is attained at
smaller driving forces and at small hold-up values. Moreover,
Almeida-Rivera45 suggested that as a heuristic for heat-transfer
processes the exchange surface per unit mass should be
maximized, while driving force is kept small at constant transfer
duty.
56 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008

Table 2. Utopia Point in Optimization Problems with More Than Table 3. Wilson Interaction Parametersa for the System
One Objective Function iC4-MeOH-MTBE-nC4 at 11 × 105 Pa5,74
Utopia point U in a RΦ-space is given by the Φ-objective function values iC4 MeOH MTBE nC4
that solve independently the Φ scalar optimization problems.62 For
aij
instance, in an optimization problem involving two objective functions
iC4 0.00000 -0.74200 0.24130 0.00000
the mathematical formulation of the utopia point U is
MeOH 0.74200 0.00000 0.98330 0.81492
min f1(x) ) flow f2(x*) ) fup MTBE -0.2413 -0.98330 0.0000 0.00000
1 (x*) and 2 (x*) (43)
x∈Γ nC4 0.00000 -0.81492 0.00000 0.00000
min f2(x) ) flow
2 (x*) and f1(x*) ) fup
1 (x*) (44) bij (K)
x∈Γ
iC4 0.0000 -85.5447 30.2477 0.0000
where f low
i (x*)and f up
i (x*) are the lower and upper limits of
the MeOH -1296.719 0.0000 -746.3971 -1149.280
objective function fi, respectively, and the set of constraints
defines the MTBE -136.6574 204.5029 0.0000 0.0000
space of design variables x. The coordinates of the utopia point are then nC4 0.0000 -192.4019 0.0000 0.0000
given by (f low low
1 , f2 )
W,i ) ∑j)1 xj × Λij; ln(γi) ) 1 - ln(SW,i) -
a ln(Λ ) ) a + b /T; S j)nc
ij ij ij
Approach To Design Optimization and References Cases. ∑k)1
k)nc
(xk × Λki/SW,k).
Two approaches are adopted to solve the optimization problem Table 4. Equilibrium Constant and Kinetic Constant for the
involving more than one objective function, Synthesis of MTBE58

( ) ()
(a) Each objective function is used independently, disregard-
1 1 T
ing the other two; this approach localizes the coordinates of ln Keq ) ln Keq,0 + Rk - + βk × ln + γk × (T - T0) + δk ×
T T0 T0
the utopia point (cf. Table 2) and, subsequently, reduces the
(T2 - T02) + k × (T3 - T03) + φk × (T4 - T04) (45)
computational effort by limiting the search space,
(b) The interactions and trade-offs among the objectives are parameters Rk ) -1.49277 × 103; βk ) -7.74002 × 101;
γk ) 5.07563 × 10-1; δ ) -9.12739 × 10-4;
accounted; in this approach, a multiobjective problem is
k ) 1.10649 × 10 ; φk ) -6.27996 × 10-10;
-6
formulated by using a weighted p-norm for the vector of T0 ) 298.15 K; Keq,0 ) 284
objective functions, as given by Clark and Westerberg,47 validity heterogeneous catalyst (Amberlyst 15)

min F ≡ [
D∈Γ
∑i (wifi)p]1/p p ∈ [1, ∞) (30) [ (
kf(T) ) kf(T*) × exp -
Ea 1
-
R T T*
1
)] (46)

parameters kf (T*) ) 15.5 × 10-3 mol × (s × eq[H+])-1;

Ea ) 92400 J × mol-1; T* ) 333.15 K
where a norm of p e 2 is used to localize noninferior points in validity heterogeneous catalyst (Amberlyst 15)
the nonconvex regions.47
This theory will now be applied to the multiobjective design Table 5. Nominal Values in the MTBE Synthesis Nominal Values in
cases for a methyl tert-butyl ether (MTBE) column. the MTBE Synthesisa
parameter value
Applications number of trays 15 [2 rectifying, 8 reactive, 5 stripping]
feed 1
(1) Steady-State Entropy Production Profile in a MTBE location iC4 + nC4, 10th tray (g)
Reactive Distillation Column. The esterification of isobutene flow rate 455 mol s-1
and methanol is used as a case study in this application. MTBE composition iC4, 0.37; nC4, 0.63
is produced from methanol (MeOH) and isobutene (iC4) in an temperature 350 K
feed 2
equilibrium-limited, liquid-phase reaction48,49 catalyzed hetero-
location MeOH, 9th tray (l)
geneously by bentonites, zeolites, or strong acidic macroporous flow rate 168 mol s-1
ion-exchange resins or homogeneously by sulfuric acid accord- composition MeOH, 1.00
ing to the following stoichiometry, temperature 320 K
operating pressure 11 × 105 Pa (top)
C4H8 + CH3OH h C5H12O (31) reflux ratio 7
mass of catalyst/tray 204.1 kg ) 1000 eq[H+]
isobutene methanol MTBE extent of reaction 150 mol s-1
column diameter 3.573 m
FCC is a commonly used feedstock for the MTBE production.49 condenser area 4.460 × 101 m2
This feedstock stream contains a complete range of butanes and reboiler area 2.456 × 102 m2
butenes. For the sake of simplicity and provided that the reaction a Operational conditions are those reported in Hauan et al.,8 Jacobs and
is highly selective for iC4, all the organic components in the Krishna,4 and Seader and Henley.59
iC4 stream are lumped together as inert nC4.5
The chemical reaction is represented by a pseudohomoge- decade.54-58 In this study, we use the model reported by Thiel
neous kinetics as reported in the literature,17,50,51 et al.58 and listed in Table 4.

( )
aiC4
aMTBE
grx(x,T) ) - (32)
aMeOH Keq × aMeOH2
In this study, we use the γ - φ thermodynamic formulation.52
The vapor phase is assumed to behave ideally, whereas the liquid
phase is modeled using the Wilson activity coefficient equa-
tion.53 The interaction parameters for the MTBE system at a
given operation condition are listed in Table 3.
The reaction equilibrium and kinetics of MTBE synthesis
have been the subject of intensive research during the past
The GLRDVEs are integrated to a column structure in such
a way that the superstructure is comparable to the unit studied
elsewhere4,8,59 and specified in Table 5. In this column structure,
the catalyst load is evenly distributed along the reactive trays4,8,59
and all the sections of the column have a common value. The
structure to be further considered during the course of this
contribution is depicted schematically in Figure 2.
To prepare for the optimization task and to reduce the problem
complexity, we identify the dominant contributions to entropy
production rate over the MTBE column. The expressions to
compute all contributions in the GLRDVE are the terms of eq
 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 57

Table 6. Optimized Design of a RD Column for MTBE Synthesis

Based on Economic Performance
parameter value
number of trays 15 [2 rectifying, 8 reactive, 5 stripping]
feed 1
location iC4 + nC4, 10th tray (g)
flow rate 455 mol s-1
composition iC4, 0.37; nC4, 0.63
temperature 350 K
feed 2
location MeOH, 9th tray (l)
flow rate 168 mol s-1
composition MeOH, 1.00
temperature 320 K
operating pressure 11 × 105 Pa (top)
reflux ratio 7
mass of catalyst/tray 204.1 kg )1000 eq[H+]
bottom flow rate 197 mol s-1
Figure 2. Schematic representation of a RD column as the integration of column diameter 7.132 m
GLRDVEs. Numbers represent tray/compartment location. condenser area 1.059 × 103 m2
reboiler area 6.93 × 102 m2
operational constraints P e Pcond
Treb < Tc,iC4
8. These expressions are applicable for all reactive and nonre- product specifications xDMTBE e 0.1%
active compartments. The reboiler and condenser models assume xBMTBE g 99%
that the entropy contribution by external heat transfer is
dominant.
The reference design (E-design) is the column structure difference of chemical potential and temperature between the
obtained by steady-state optimization of the spatial design entering streams and the mixture coexisting on those stages,
variables with respect to the unit’s economic performance. For (d) External heat transfer plays a dominant role in the reboiler
this optimization problem, the following formulation is con- and condenser, exclusively.
sidered,
In terms of entropy production, summarizing, the separation
Function: min fecon ) TAC (33) function dominates over the reaction function for this MTBE
Dcol,Areb,Acond∈Γ
RD column configuration.
s.t. gx(‚) ) 0 (2) Biobjective Optimization of a MTBE Reactive Distil-
lation Column. A multiobjective optimization problem is
hx(‚) e 0
formulated for the RD column with respect to economic
D
xMTBE D
e xMTBE,spec performance and exergy efficiency (see Table 7). The formula-
tion includes the balances described by eqs 1-5 and 18, the
B
xMTBE B
g xMTBE,spec criteria definitions given in eqs 29, and the optimization
formulation described by eq 30. Constraints imposed by
where the path constraints hx(‚) must be satisfied at all times operating conditions and product specifications are included.
during the system operation.60,61 For the sake of building up the optimization complexity, the
The relevant parameters and schematic representation of the first approximation of this approach omits the response time
optimized design are given in Table 6 and Figure 2, respectively. constant as objective function.
The entropy profiles along the column for this reference The resulting optimization problem is formulated as follows
design are presented in Figure 3 and calculated according to eq and is implemented and solved in gPROMS/gOPT,
8. The following results are of interest,
(a) The entropy profiles show that energy transfer contributes
largely to the overall entropy production and particularly in the
stripping section of the column, followed by mass diffusion. In
the reactive zone, however, the dominant source of entropy
production is chemical reaction. In the rectifying section, both
mass and energy-transfer contributions are equally important,
(b) The entropy contribution originated from chemical
reaction shows negative values at the bottom of the reactive
section. This fact may be explained by possible inconsistency
between the reaction kinetics model and the affinity expression.
Although MTBE decomposition takes place at that column
location, due to high MTBE concentration, the reaction affinity
is excepted also to change sign exactly at the same point in the
(P,T,x) space. These facts result always in a nonnegative entropy
production contribution. Moreover, second law requires only
that the sum of all contributions to the entropy production is Figure 3. Entropy production rate profile for a 15-stage RD column for
MTBE synthesis. Stages are numbered from top to bottom; reboiler and
larger than zero, condenser drum are not depicted; MeOH feed stream is fed at 9th tray and
(c) At the feed stages (9th-10th), the conjugated driving iC4 stream at 10th tray. The operational and design variables are listed in
forces are considerably larger than elsewhere due to the Table 6.
58 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008

Table 7. Optimized Design of a RD Column for MTBE Synthesis

Based on Economic Performance and Exergy Efficiencya
parameter value
number of trays 15 [2 rectifying, 8 reactive, 5 stripping]
feed 1
location iC4 + nC4, 10th tray (g)
flow rate 455 mol s-1
composition iC4, 0.37; nC4, 0.63
temperature 350 K
feed 2
location MeOH, 9th tray (l)
flow rate 168 mol s-1
composition MeOH, 1.00
temperature 320 K
operating pressure 11 × 105 Pa (top)
reflux ratio 7
mass of catalyst (eq[H+]) m1:2 ) 0; m3 ) 771.9
m4 ) 826.122; m5 ) 762.405
m6 ) 688.412; m7 ) 652.292
m8 ) 785.991; m9 ) 1359.42
m10 ) 982.448; m11:15 ) 0
bottom flow rate 197 mol s-1 Figure 4. Pareto optimal curve f ′econ versus f ′exergy. Each point of the
column diameter curve represents an optimal solution in terms of the design variables [mcat,
rectifying 6.95 m Q, aΨ] and weighting factors w; axes are normalized so that utopia point
reactive 6.95 m coordinates are as follows: U ) (0,0); original coordinates of U are: U0 )
stripping 6.33 m (5.251 × 104 J × (s × K)-1,2.082 × 108 euro × y-1).
condenser area 1.059 × 103 m2
reboiler area 6.93 × 102 m2
a w ) [0.3 0.7]T.

min J(fecon(x),fexergy(x),w) (34)

D∈Γ

s.t. gx(‚) ) 0
hx(‚) e 0

where the set of design variables D include variables as ([mcat

Q aΨ]). Operational variables such as reflux ratio and condenser
pressure have been fixed based on the explanation given in
Design Decision Variables at appropriate values as listed in
Table 6. The inequality constraints hx(‚) are defined by operating
conditions and product specifications (cf. bottom section of
Table 6).
The planar view of the two normalized objective functions
at different weighting factors is depicted in Figure 4. The
normalization of the objectives requires a priori knowledge of
the minimum (fmin) and maximum (fmax) values for each
objective. These values are found by solving the optimization
problem at different weighting factors. Each objective value is
then scaled based on the corresponding utopia coordinate and
for a given weighting factor. Thus, the normalized objective
functions (f′) are computed according to the expression,
Figure 5. Normalized catalyst distribution in the synthesis of MTBE with
respect to economic performance and exergy efficiency.
deviations between optimal designs and the utopia point (up to
60%) for the pair f ′econ - f ′exergy.66 The decision maker should
have in mind the importance of compromising both objectives
at early stages of the design cycle.
Significant structural differences exist between the economic-
driven (E) and exergy-driven (X) designs, which are respectively
given as

hf i - hf min E; w ) [1 0 ]T
i
f′i ) (35)
hf max
i - hf min
i
and

where the objective function hfi is scaled with respect to the X; w ) [0 1 ]T

corresponding coordinates of the utopia point, As depicted in Figure 5, for instance, the catalyst load along

( ) fi p the reactive trays differs among the proposed designs when

hf i ) wi × i ∈ Znp
fUi
The curve of Figure 4 is termed Pareto curve and clearly
shows that one objective function can only be improved at the
expense of the other objective function.1,62 The economics-
exergy trade-off has been extensively studied in energy-related
literature27,63,64 and in the engineering branch of exergoeco-
nomics.65 In our RD case study, this trade-off leads to significant
(36) compared to the reference case. As expected, the optimal design
based on economic performance (E) coincides with the nominal
case (R). The catalyst load of the nominal case is used to
normalize the plot. When exergy is fully considered as an
objective function, the last reactive tray vanishes together with
the sixth reactive tray. Furthermore, the catalyst load of the
seventh reactive tray is considerably higher than that of the
nominal case. This output suggests an alternative design, where
a side reactor with high catalyst load is placed at the seventh

Figure 6. Entropy production rate profile for an optimal design of a MTBE
RD column based on exergy efficiency (X-design). Stages are numbered
from top to bottom; MeOH feed stream is fed at 9th tray and iC4 stream at
10th tray.
tray. The catalyst load on the remaining reactive trays decreases
when exergy is taken into account. The design with equally
weighted objectives (Q; w ) [0.5 0.5]T) presents an interesting
catalyst load profile, specially at the ninth tray. The rather large
catalyst load might be explained by the reduced catalyst load
on the last reactive tray and the constraint of targeting the
product specifications. The reduced catalyst load on the ninth
tray diminishes the MTBE decomposition at high MTBE
composition and the product synthesis at low MTBE composi-
tion. As a constraint in the product specification is included in
the problem formulation, the required extent of reaction is
accommodated only by having the ninth tray fully loaded with
catalyst. Since MeOH is fed at the ninth tray, larger holdup of
reactant is present and only converted into products by an
increased amount of catalyst. The effect of catalyst amount on
the performance of a RD column has also been addressed,67
aiming at further internal heat integration between the reaction
and separation tasks.
When the entropy profiles of the X-design (cf. Figure 6) are
compared to those of the E-design, it can be realized that the
mayor contribution of entropy is still coming from the heat-
transfer process in the stripping section. However, mass diffusion
becomes more relevant than in the E-design. In the reactive
zone and in contrast to the results of Figure 3, the exergy by
chemical reaction is unevenly produced along the reactive trays.
Thus, the ninth tray largely contributes to the overall exergy
produced within the reactive section. A fully catalyst load on
that tray intuitively explains this behavior.
According to these last results, for the exergy-driven design
(X-design), it seems more attractive to reach high conversions
on one or a few trays and low conversion on surrounding trays.
Apparently, this finding seems to disagree with the theorem of
equipartition of driving forces.68,69 As demonstrated by Johan-
nessen and Rosjorde40 and mentioned by Jimenez et al.,63 this
theorem is a good approximation to the state of minimum
entropy production in an optimally controlled system, only if
the system has sufficient freedom. For instance, in conventional
distillation equipartition of driving forces results in a diabatic
separation column with minimum total entropy production at
large areas and at large number of trays.40,63
(3) Triobjective Optimization of a MTBE Reactive Distil-
lation Column: A Sensitivity-Based Approach. For the sake
of simplification, the following strategy is adopted: (i) an
optimized design based in economics and exergy efficiency is
considered; (ii) a disturbance scenario is defined; and (iii) the
Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 59
Table 8. Entropy Produced in Classical and Green Designsa
w fecon (euro y-1) fexergy (J (s K)-1)
[1.0 0.0]T 2.082 × 108 5.604 × 104
[0.3 0.7]T 2.085 × 108 5.305 × 104
[0.0 1.0]T 2.086 × 108 5.251 × 104
a Classical design corresponds to w ) [1 0]T; green design corresponds
to w ) [0.3 0.7]T.
responsiveness index of the optimized design is estimated at
the given disturbance scenario.
Optimized Design. According to applications 1 and 2, the
optimized designs based on fecon and fexergy present an important
tradeoff between both objectives. This tradeoff is dependent on
the vector of weighting factors (w). The ultimate decision on w
is made by the designer, according to the requirements of the
design problem. In our case, and only for tutorial purposes, we
choose the vector
w ) [0.3 0.7 ]T
over all available designs. The optimized design variables for
this case are listed in Table 8.
Disturbance Scenario. The selection of a disturbance
scenario is carried out by analyzing qualitatively the entropy
profiles of Figure 3 (E-design) and Figure 6 (X-design) and by
focusing on process disturbances occurring in industrial practice.
An upstream variation of the MeOH feed flow rate up to -10%
is chosen as disturbance scenario.
FMeOH ) FSS
MeOH × (1 - Rτ(t))
(for 0 e Rτ(t) e 0.1, t ∈[0,tf]) (37)
The disturbance scenario is chosen so that any steady-state
transition is avoided.
Control Structure. To mitigate any effect caused by a
disturbance scenario, a control structure is proposed for the
RD unit shown in Figure 2. The following assumptions
are considered: (i) four control valves are associated with the
RD unit; (ii) the unit is operated at constant reflux ratio RR
() FR,L/FD,L ) 7); and (iii) the throughput of the unit is set.
Based on a rigorous degree of freedom analysis45 a 4 × 4 SISO
control structure is suggested. The controlled variables are
chosen aiming at the satisfaction of product specifications and
protection of the unit. The pairing of controlled-manipulated
variables, on the other hand, tries to minimize cross-interactions
between manipulations of input variables on the controlled
output variables and looks for a dominant 1-1 correspondence
between input and output variables. Thus, two control valves
are used to control the liquid levels in the reflux drum and
column bottom. As the reflux ratio is larger than a value of 4
(RR > 4), the reflux drum level is controlled with the liquid
reflux or drum liquid outflow (Richardson’s rule31). The level
at the column bottom might be by convention controlled with
the reboiler liquid outflow. The third control valve is used to
manipulate the flow rate of cooling utility to control the column
pressure. The remaining control degree of freedom is used to
control the purity of the MTBE product by manipulating the
reboiler heat duty. The composition of MTBE at the bottom
stream is conveniently inferred from temperature measurements
at a given tray. Steady-state open-loop simulations are carried
out to assess the effect of varying manipulated variable (i.e.,
reboiler duty Qreb). The simulation results indicate45 that the
largest temperature changes are occurring either at the rectifying
section, the top section of the reactive section or at the reboiler.
By selecting the reboiler temperature to infer the MTBE
60 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008

Responsiveness Index. For each MeOH feed rate, the

Figure 7. Schematic representation of a MTBE RD column with a 4 × 4
SISO control structure. Numbers represent tray location.
composition in the bottom stream, any hydraulic lag is ef-
fectively avoided.
This single-end point composition control structure is origi-
nated from the assumption of invariant reflux ratio. As
mentioned by Luyben et al.,31 this control structure is more
advantageous from both operational (i.e., easy tuning) and
economical (i.e., comparable energy demand) points of view
than the dual composition control. Figure 7 depicts the pro-
posed control structure. The controllers are tuned by solving
an optimization problem in the time domain, aiming at a
minimization of the integral absolute error of the control
loops,70
variables B(x,u) and ΣV(x,u) are estimated in the corresponding
steady state and the responsiveness index is computed. The
responsiveness index is treated as a function of the magnitude
of the MeOH feed perturbation, fresponse(FMeOH).
Specifically, this MeOH flow rate disturbance results in a
variation of the mass-transfer driving forces for each component
along the unit. From this point on, the analysis is focused on
the top tray (1st), the 9th tray, and the bottom tray (15th) of the
RD column specified in Table 8. Considering these tray
locations, we provide a generalized overview on the unit’s
responsiveness in the rectifying, reactive and stripping sections.
The 9th tray is chosen because it is on that particular feeding
tray where MeOH is fed to the column and, therefore, where a
more pronounced effect of feed flow rate disturbance is
expected.
The corresponding mass-transfer and chemical reaction
driving forces are plotted for the 1st, 9th, and 15th trays in
Figure 8. It can be seen from these plots that both MeOH mass-
transfer driving forces increase with larger disturbances, whereas
chemical reaction driving force decreases.
The response time of the MTBE RD column as a function of
Xm,MeOH (1st and 15th trays) and Xrx (9th tray) is depicted in
Figure 9. The variations of B(x,u) and ΣV(x,u) are computed
between the initial reference state and between consecutive
states. The response time becomes larger at small Xm,MeOH-values
in the rectifying and stripping sections, whereas small driving
forces lead to short response times in the reactive section. From
a responsiveness perspective, therefore, small driving forces are
desirable for reactive sections and large driving forces are
desirable for separation (nonreactive) sections.
The first statement suggests that in the case of reactive
sections driving forces should be manipulated to control the
unit while flux is specified (i.e., a certain conversion rate). To
explain this statement. we require the introduction of an
approximated relation between fluxes and forces for a kth
compartment,31,73

Function: min
Ki,Bi,τc,i,SPi
F) ∫0t [(xb - xjb)2 + (xd - xjd)2] dt
f
Jk ≈ R × Xk (39)

s.t. fx(‚) ) 0 t ∈ [t0,tf] T0 × ∆ΣV ≈ 2 × T0 ∫V dV ∑ R × Xk × ∆Xk (40)

k
gx(‚) ) 0 t ∈ [t0,tf]
hx(‚) e 0 t ∈ [t0,tf] where R is a design factor, which Meeuse called the thermo-
dynamic design factor.46
dx(‚) ) 0 t ∈ [t0,tf] At small gradient X, the design factor R should be large
enough to obtain the required flux. In the reactive section, small
ix(‚) e 0 t ∈ [t0,tf] reaction driving forces require large reaction volumes on the
trays. Small variations in the MeOH feed flow rate produce
Tmin max
reb e Treb e Treb t ∈ [t0,tf] large changes in the affinity and subsequently large entropy
production rate values. As the stored exergy changes only
Tout,min out out,max
water e Twater e Twater t ∈ [t0,tf]
modestly with a small flow disturbance, the overall effect
Tout,min out out,max
t ∈ [t0,tf] enhances the process responsiveness (i.e., small response times).
steam e Tsteam e Tsteam (38)
The second statement can be explained by following the same
where the objective function F accounts for deviations from reasoning. In a distillation (separation) process, the driving force
product and top specifications (xj) and the constraint paths set (top-bottom) is specified and the flux remains to influence.
bounds for the temperature in the reboiler and the outlet Thus, in nonreactive sections, flux should be manipulated to
temperatures of the cold and hot utilities. control the unit. The results of these observations fully agree
Alternative control configurations are proposed for the MTBE with the analysis given by Luyben et al.31 and Meeuse46 for a
synthesis by Wang et al.71 and Sneesby et al.72 addressing reactor and distillation column.
switchover to different steady states. As the focus of this Meaningfulness and Validity of Results. In the current
contribution is the dynamic behavior around a known steady approach, a continuous range of disturbances for one input
state, no steady-state transition is considered by constraining variable is considered for the computation of the responsiveness
the disturbances’ variability. time constant. In comparison with the generalized definition of
 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 61

Figure 8. Driving forces as a function of the MeOH feed flow rate. MeOH feed flow rate varies from the nominal value (168 mol s-1) to 150 mol s-1; the
optimal design variables and sizing parameters corresponds to a fecon - fexergy design, with w ) [0.3 0.7]T.

Figure 9. Response time as a function of the MeOH feed flow rate. MeOH feed flow rate varies from the nominal value (168 mol s-1) to 150 mol s-1; the
optimal design variables and sizing parameters corresponds to a fecon - fexergy design, with w ) [0.3 0.7]T.

fresponse by Luyben et al.,31 this is a very specific case. Their not optimized over the feasible domain of states for normal
(Luyben’s et al.) definition demands an optimization of the states operation. Our approach might be locally optimized for the given
x to find the worst case situation over a domain of states. operational window, but might have a low degree of validity to
Therefore, there is a latent risk that the presented results are other cases.
62 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008
Comparison between Classical and Green Designs
In the course of this contribution, the economic-driven design
(classical design) was implicitly compared to a new bioptimized
design (green design) for the case of a MTBE RD column. A
set of indicators can be used to compare both designs: (i) exergy
loss; and (ii) closed-loop control performance.
The exergy loss (or entropy produced) can be extracted from
the Pareto curve presented in Figure 4. The relevant data for
the classical and green designs and for the fully exergy-driven
design, i.e.,
w ) [0 1 ]T
are listed in Table 8.
As was expected, the green design produces less entropy than
the classical one; however, economically it is less attractive than
the conventional design, as it presents a higher fecon value. The
increase in costs is rather marginal (∼0.2%) and within the
uncertainty range of the correlation to predict investments costs
for units and catalyst. A tendency toward increasing cost can
be suggested though,66 it being not significant for this particular
case. A more significant change is observed for fexergy (∼6%).
A more revealing concept to assess the impact of making a
process greener is the shadow price λp. It is a cost penalty
function for changing entropy production and given by
∂fecon
λp ) (41)
∂fexergy
The shadow prices can be computed directly out from the
data in Table 8 or from the Pareto curve in Figure 4. It was
found that, when compared to the classical case, the green design
presents a cost penalty of approximately -2.02, whereas the
X-design presents a value of approximately -4.37 [euro × K
× MJ-1]. These values confirm the tradeoff between both
objectives and suggest that the greener the design the larger
the involved cost penalty.
The closed-loop control performance of both designs is now
analyzed. The disturbance scenario under consideration repre-
sents a step variation in the flow rate of methanol during the
time horizon of 6 × 104 s and given by,
{
FMeOH )
FSS
MeOH (t ∈ [4 × 104 s, 6 × 104 s])
if t ∈ [0 s, 1 × 104 s] or
(42)
1.1 × FSS
MeOH (if t ∈ [1 × 10 s, 4 × 10 s])
4 4
The dynamic response of the classic and green designs are
depicted in Figure 10. In this figure, the MTBE composition at
the bottom stream is presented as a function of time in the
presence of the MeOH feed disturbance. The bioptimized design
(i.e., green) has an improved closed-loop performance compared
to the classical design. The deviations from xMTBE set point are
less pronounced in the green design, which smoothly returns
back to the initial steady state when the disturbance disappears.
The set point of the green design is slightly lower (0.05%) than
the one of the classic design. This negligible variation in set
points does not affect the quality of the analysis results. It is
suggested from our results that incorporating both economic
and exergy-related objectives in the unit design results in a
process with better closed-loop performance, provided that
control structure and tuning are kept invariant.A well-tuned
control can compensate for any change in the process operability
feature.
Figure 10. Time variation of MTBE product stream for the classic and
green designs in the presence of a MeOH feed flow rate disturbance. Green
design corresponds to w ) [0.3 0.7]T; column specifications for classic
and green designs are given in Tables 6 and 8, respectively.
Concluding Remarks
The starting point of this contribution was the definition of
a generic building block for process synthesis (GLRDVE). Being
embedded in a superstructure, each GLRDVE block has its own
complexities due to the many interacting phenomena. However,
the modeling framework of the GLRDVE does not account for
the interactions between the stages in a real multistage column.
To overcome this limitation, operational variables such as reflux
ratio and condenser pressure were specified at a higher decision-
making layer aiming at operating windows where steady-state
multiplicity and reactive azeotropy phenomena are not likely
to occur. This a priori specification might be considered a
drawback of the proposed approach, but it is certainly a topic
to be further researched. The sources and the distribution of
the entropy production rate in a complete MTBE column were
then determined and analyzed. It is believed that insight into
such a distribution is helpful for assessing which column-related
design decision variables (e.g., feed tray location, reflux ratio,
heat load distribution using diabatic operation, catalyst load,
and mass/heat-transfer area) are key to optimize the column
performance. The perspective adopted for selecting the design
decision variables was task-oriented rather than the convention-
ally used equipment-oriented design approach. The chosen
design decision variables included the reaction volume, the
amount of heat to be added or removed, and the interfacial areas
for species mass and heat transfer. Analyzing the steady-state
entropy production profiles in a MTBE RD column showed that,
in terms of entropy production, the separation function domi-
nates over the reaction function. A biobjective function was then
studied regarding economic and exergy performance. It is
suggested that designs based exclusively on economic or exergy
performances differ significantly in their structure, showing an
existing trade-off. The Pareto curve provides sensitivity infor-
mation on the cost effects of improving the exergy efficiency.
In addition to economics and exergy objectives, this optimiza-
tion-based approach covers responsiveness aspects. To define
a responsiveness criterion, the availability and entropy produc-
tion rate were regarded as functions of an internal state vector
x and a vector of disturbances or manipulated external excita-
tions u. In our approach, the maximization approach suggested
by Luyben et al.31 was not considered due to its inherent need
for high computational effort. However, a rather formal defini-
tion of fresponse was adopted together with a single input
excitation. By not searching over all possible state trajectories,
 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 63

the generality of the results might be questionable and the Xm,j ) conjugated driving forces for interfacial mass diffusion
external excitation is a fairly conservative one. By analyzing of component j
the responsiveness criteria of the bioptimized design, it is XHX ) conjugated driving forces for heat exchange with the
reconfirmed that small driving forces are desirable for reactive external world
sections and large driving forces are desirable for separation Xrx,j ) conjugated driving forces for the chemical reaction of
(nonreactive) sections.31,46 The first statement suggests that, in reaction j
the case of reactive sections, driving forces should be manipu- hf ) scaled objective function
lated to control the unit. The second statement suggests that in xj ) product specification
nonreactive sections flux should be manipulated to control the E ) economic-based optimal design
unit. Finally, the economic-driven design (classical design) was Q ) optimal design based equally on economic performance
compared to the new bioptimized design (green design) for the and exergy efficiency
case of the MTBE RD column. The green design produced less U ) Utopia point
entropy than the classical one, while being marginally less X ) exergy-based optimal design
attractive from an economic standpoint. The bioptimized design A Li ) affinity of chemical reaction i in the liquid phase
had an improved closed-loop performance compared to the D ) set of design or decision variables
classical design, keeping the same control structure and settings. R ) universal gas constant
It is concluded that incorporating both economic and exergy- A ) availability
related objectives in the unit design results in a process with B ) availability function
better closed-loop performance and reduced exergy loss. B(x,u)total ) availability over all the superstructure
In this contribution, we have focused on the performance of aΨ ) mass and heat transfer area
RD process designs, paying less explicit attention to the ai ) activity of component i
industrial conventional reaction-separation processing route. aij, b ij ) Wilson model parameters
Therefore, a comparison between RD and non-RD processes c ) investment cost of unit and catalyst; cost of component or
with regard to entropy production and the design implications utility per molar unit
for the principle of equipartition of forces are recommended dx(‚) ) set of disturbances
topics to be further researched. f ) objective function
f′ ) normalized objective function
Nomenclature fecon ) economic performance indicator
fexergy ) sustainability performance index
List of Symbols
fresponse ) responsiveness performance index
A ) area fx(‚) ) set of differential expressions
B ) controller’s bias gj,(R) ) external force exerted on component i per molar unit in
i
Dcol ) column diameter stream j and phase (R)
total ) total flow of stream j in the vapor phase; Ftotal
F j,V j,V )
gx(‚) ) set of algebraic expressions
i)nc
Σ i)1 Fij,V grx(‚) ) chemical reaction driving force based on activities
F j,(R)
i ) convective flow of component i in phase R in the hj,(R)
i ) molar enthalpy of component i in stream j and phase
incoming or outgoing stream (R)
i ) convective flow of component i in compartment k in the
F j,k hx(‚) ) set of inequalities and path constraints
incoming or outgoing ix(‚) ) set of initial conditions
stream j for the exchange with the external world m ) mass
e ) total energy flux in phase (R)
J (R) nc ) number of components
J Lq ) measurable heat flux into the liquid phase no ) number of objective functions
np ) number of phases
s ) entropy flux in phase (R)
J (R)
nt ) number of compartments in the integrated structure
(R)
J m,i ) mass flux of component i in phase (R)
nrx ) number of chemical reactions
K ) controller’s gain nst ) number of incoming and outgoing streams
Keq ) temperature-dependent chemical equilibrium constant p ) norm
M(R) ) molar holdup in phase (R) q(R′ - R) ) heat flux from phase (R′) to phase (R)
P ) absolute pressure ri(R) ) net change of the amount of component i in phase (R)
Q(R) ) external heat exchange rate in phase (R) due to all reactions
RR ) reflux ratio s(R) ) entropy density in phase (R)
SW ) variable in Wilson model t ) time
Se ) entropy produced by interfacial energy transfer u ) internal energy density of the volume element
S j,(R)
i ) molar entropy of component i in stream j and phase w ) vector of weighting factors
(R) x ji ) liquid molar fraction of component i in stream j
Sm ) entropy produced by interfacial mass diffusion y ji ) vapor molar fraction of component i in stream j
Srx ) entropy produced by chemical reaction R ) nominal design
T(R) ) absolute temperature in phase (R) U(R) ) internal energy of phase (R) or compartment
V ) volume of compartment
V LRX ) reaction volume in the liquid phase Acronyms
Wideal ) minimum work required by the process nC4 ) n-butane
Xq ) conjugated driving forces for heat exchange between iC4 ) isobutene
phases GLRDVE ) generic lumped reactive distillation volume element
Xe ) conjugated driving forces for interfacial heat diffusion MeOH ) methanol
64 Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008

MTBE ) methyl tert-butyl ether x ) state variables

RD ) reactive distillation
SP ) set point Literature Cited
TAC ) total annualized cost
(1) Hakanen, J.; Hakal, J.; Manninen, J. An Integrated Multiobjective
F ) multiobjective function Design Tool for Process Design. Appl. Therm. Eng. 2006, 26, 1393.
(2) Rodriguez, I. E.; Zheng, A.; Malone, M. F. Parametric Dependence
Greek Symbols of Solution Multiplicity in Reactive Flashes. Chem. Eng. Sci. 2004, 59,
R ) thermodynamic design factor 1589.
(3) Lakerveld, R.; Bildea, C. S.; Almeida-Rivera, C. P. Exothermic
δ(R) ) Kronecker delta function Isomerization Reaction in Reactive Flash: Steady-State Behavior. Ind. Eng.
ηII ) efficiency of the 2nd law of thermodynamics Chem. Res. 2005, 44, 3815.
Γ ) set of constraints (4) Jacobs, R.; Krishna, R. Multiple Solutions in Reactive Distillation
γi ) liquid activity coefficient for component i for Methyl tert-Butyl Ether Synthesis. Ind. Eng. Chem. Res. 1993, 32, 1706.
(5) Güttinger, T. E. Multiple Steady States in Azeotropic and Reactive
λp ) shadow price Distillation. Ph.D. Dissertation, Swiss Federal Institute of Technology,
Λij ) Wilson model binary interaction parameter for the pair i, Zürich, 1998.
j ∈ Znc (6) Kienle, A.; Groebel, M.; Gilles, E. D. Multiple Steady States in
Binary Distillation. Theoretical and Experimental Results. Chem. Eng. Sci.
µi j,(R)
) chemical potential of component i in stream j and 1995, 50, 2691.
phase (R) (7) Ciric, A.; Miao, P. Steady-State Multiplicites in a Ethylene Glycol
i,j ) stoichiometric coefficient of component i in reaction j
ν (R) Reactive Distillation Column. Ind. Eng. Chem. Res. 1994, 33, 2738.
and phase (R) (8) Hauan, S.; Hertzberg, T.; Lien, K. Multiplicity in Reactive Distillation
of MTBE. Comput. Chem. Eng. 1995, 19, S327.
Ψe ) energy flux associated with the interfacial mass and heat (9) Hauan, S.; Hertzberg, T.; Lien, K. Why Methyl tert-Butyl Ether
transfer Production by Reactive Distillation May Yield Multiple Solutions. Ind. Eng.
Ψi ) interfacial molar flux of component i Chem. Res. 1995, 34, 987.
(10) Güttinger, T. E.; Morari, M. Predicting Multiple Steady States in
F(R)
i ) molar density of component i in phase (R) Distillation: Singularity Analysis and Reactive Systems. Comput. Chem.
ΣV ) total entropy production rate in phase (R)
(R)
Eng. 1997, 21, S995.
ΣV(x,u) ) total rate of entropy production over all superstructure (11) Güttinger, T. E.; Morari, M. Predicting Multiple Steady States in
Equilibrium Reactive Distillation. 2. Analysis of Hybrid Systems. Ind. Eng.
V ) entropy production rate per unit of volume in phase (R)
σ (R) Chem. Res. 1999, 38, 1649.
τ ) settling or response time (12) Güttinger, T. E.; Morari, M. Predicting Multiple Steady States in
τc ) controller’s response time Equilibrium Reactive Distillation. 1. Analysis of Nonhybrid Systems. Ind.
Eng. Chem. Res. 1999, 38, 1633.
Li ) degree of advancement of chemical reaction i in liquid (13) Higler, A.; Taylor, R.; Krishna, R. Nonequilibrium Modelling of
phase Reactive Distillation: Multiple Steady States in MTBE Synthesis. Chem.
Eng. Sci. 1999, 54, 1389.
Subscripts and Superscripts (14) Chen, F.; Huss, R. S.; Doherty, M.; Malone, M. F. Multiple Steady
States in Reactive Distillation: Kinetic Effects. Comput. Chem. Eng. 2002,
0 ) initial value; infinitely large sink with similar surrounding
26, 81.
conditions (15) Sneesby, M. G.; Tade, M. O.; Smith, T. N. Mechanistic Interpreta-
U ) Utopia point tion of Multiplicity in Hybrid Reactive Distillation: Physically Realizable
S(R) ) entropy of compartment in phase (R) Cases. Ind. Eng. Chem. Res. 1998, 37, 4424.
(16) Sneesby, M. G.; Tade, M. O. R. S. Multiplicity and Pseudo-
B ) bottom stream
multiplicity in MTBE and ETBE Reactive Distillation. Chem. Eng. Res.
cu ) cool utility Des. 1998, 76, 524.
hu ) hot utility (17) Mohl, K. D.; Kienle, A.; Gilles, E. D.; Rapmund, P.; Sundmacher,
D ) distillate stream K.; Hoffmann, U. Steady-state Multiplicities in Reactive Distillation
Columns for the Production of Fuel Ethers MTBE and TAME: Theoretical
e ) energy-based Analysis and Experimental Validation. Chem. Eng. Sci. 1999, 54, 1029.
f ) final (18) Malone, M. F.; Doherty, M. F. Reactive Distillation. Ind. Eng.
HX ) external heat exchange Chem. Res. 2000, 39, 3953.
m ) mass-based (19) Barbosa, D.; Doherty, M. Design and Minimum-reflux Calculations
for Single-feed Multicomponent Reactive Distillation Columns. Chem. Eng.
Q ) heat source or heat sink Sci. 1988, 43, 1523.
s ) entropy-based (20) Song, W.; Huss, R. S.; Doherty, M.; Malone, M. F. Discovery of
cat ) catalyst Reactive Azeotropes. Nature 1997, 388, 561.
in ) inlet conditions (21) Frey, T.; Stichlmair, J. Reactive Azeotropes in Kinetically Con-
trolled Reactive Distillation. Trans. Inst. Chem. Eng., Part A 1999, 77, 613.
low ) lower bound (22) Harding, S. T.; Floudas, C. A. Locating All Heterogeneous and
max ) maximal value Reactive Azeotropes in Multicomponent Mixtures. Ind. Eng. Chem. Res.
min ) minimal value 2000, 39, 1576.
(23) Maier, R.; Brennecke, J.; Stadtherr, M. Reliable Computation of
out ) outlet conditions Reactive Azeotropes. Comput. Chem. Eng. 2000, 24, 1851.
reb ) reboiler (24) Frey, T.; Stichlmair, J. Thermodynamic Fundamentals of Reactive
spec ) specification value Distillation. Chem. Eng. Technol. 1999, 22, 11.
SS ) steady state (25) Ung, S.; Doherty, M. Necessary and Sufficient Conditions for
Reactive Azeotropes in Multireaction Mixtures. AIChE J. 1995, 41, 2383.
total ) total value or combined stream (26) Apaiah, R.; Kinnemann, A.; vander Kooi, H. J. Exergy Analysis:
u ) utility A Tool to Study the Sustainability of Food Supply Chains. Food Res. Int.
up ) upper bound 2006, 39, 1.
(27) Rivero, R.; Garcia, M.; Urquiza, J. Simulation, Exergy Analysis
Vectors and Application of Diabatic Distillation to a Tertiary Amyl Methyl Ether
Production Unit of a Crude Oil Refinery. Energy 2004, 29, 467.
g(R) ) external forces in phase (R) (28) Koeijer, G.; Rivero, R. Entropy Production and Exergy Loss in
u ) disturbances or manipulated excitations Experimental Distillation Columns. Chem. Eng. Sci. 2003, 58, 1587.

(29) Takeshita, K.; Ishida, M. Optimum Design of Multi-stage Isotope
Separation Process by Exergy Analysis. Energy 2006, 31, 3097.
(30) Iwakabe, K.; Nakaiwa, M.; Huang, K.; Nakanishi, T.; Rosjorde,
A.; Ohmori, T.; Endo, A.; Yamamoto, T. Energy Saving in Multicomponent
Separation Using an Internally Heat-integrated Distillation Column (HIDiC).
Appl. Therm. Eng. 2006, 26, 1362.
(31) Luyben, W. L.; Tyreus, B. D.; Luyben, M. L. Plantwide Process
Control; McGraw-Hill: New York, 1999.
(32) Kjelstrup, S.; Bedeaux, D. Elements of IrreVersible Thermodynam-
ics; International Centre for Applied Thermodynamics: Istanbul, 2001.
(33) Marquardt, W. Modeling and Analysis of Chemical Process
Systems. Presented at Delft University of Technology, Delft, The Nether-
lands, July 2004.
(34) Bird, B.; Stewart, W.; Lightfoot, E. Transport Phenomena; John
Wiley and Sons, Inc.: New York, 2003.
(35) Seider, W.; Seader, J. D.; Lewin, D. Process Design Principles;
John Wiley and Sons, Inc.: New York, 1999.
(36) Bedeaux, D.; Kjelstrup, S. Irreversible Thermodynamics. A Tool
to Describe Phase Transitions Far from Global Equilibrium. Chem. Eng.
Sci. 2004, 59, 109.
(37) Koeijer, G. Energy Efficient Operation of Distillation Columns and
a Reactor Applying Irreversible Thermodynamics. Ph.D. Dissertation,
Norwegian University of Science and Technology, Trondheim, Norway,
2002.
(38) Kjelstrup, S. Norwegian University of Science and Technology at
Trondheim, Trondheim, Norway. Personal communication, 2005.
(39) Koeijer, G.; Kjelstrup, S.; vander Kooi, H. J.; Gros, B.; Knoche,
K. F.; Andersen, T. R. Positioning Heat Exchangers in Binary Tray
Distillation Using Isoforce Operation. Energy ConVers. Manage. 2002, 43,
1571.
(40) Johannessen, E.; Rosjorde, A. Equipartition of Entropy Production
as an Approximation to the State of Minimum Entropy Production in
Diabatic Distillation. Energy 2007, 32, 467.
(41) Ismail, S.; Proios, P.; Pistikopoulos, E. Modular Synthesis Frame-
work for Combined Separation/Reaction Systems. AIChE J. 2001, 47, 629.
(42) Papalexandri, K.; Pistikopoulos, E. Generalized Modular Repre-
sentation Framework for Process Synthesis. AIChE J. 1996, 42, 1010.
(43) Bermingham, S. A Design Procedure and Predictive Models for
Solution Crystallisation Processes. Ph.D. Dissertation, Delft University of
Technology, Delft, The Netherlands, 2003.
(44) Coulson, J. M.; Richardson, J. F. Chemical Engineering Design;
Chemical Engineering, Vol. 6; Butterworth-Heinemann: Oxford, Boston,
1997.
(45) Almeida-Rivera, C. P. Designing Reactive Distillation Processes
with Improved Efficiency. Ph.D. Dissertation, Delft University of Technol-
ogy, Delft, The Netherlands, 2005.
(46) Meeuse, M. On the Design of Chemical Processes with Improved
Controllability Characteristics. Ph.D. Dissertation, Delft University of
Technology, Delft, The Netherlands, 2003.
(47) Clark, P.; Westerberg, A. Optimization for Design Problems Having
More than One Objective. Comput. Chem. Eng. 1983, 7, 259.
(48) Jimenez, L.; Wanhschafft, O.; Julka, V. Analysis of Residue Curve
Maps of Reactive and Extractive Distillation Units. Comput. Chem. Eng.
2001, 25, 635.
(49) Peters, U.; Nierlich, F.; Schulte-Körne, E.; Sakuth, M. Methyl tert-
Butyl Ether. In Ullmann’s Encyclopedia of Industrial Chemistry; John Wiley
and Sons, Inc.: New York, 2000.
(50) Sundmacher, K.; Hoffmann, U. Multiple Reactions in Catalytic
Distillation Processes for the Production of the Fuel Oxygenates MTBE
and TAME: Analysis by Rigourous Model and Experimental Validation.
Chem. Eng. Sci. 1999, 54, 2839.
(51) Schrans, S. M.; Wolf, S.; Baur, R. Dynamic Simulation of Reactive
Distillation: An MTBE Case Study. Comput. Chem. Eng. 1996, 20, S1619.
(52) Malanowski, S.; Anderko, A. Modelling Phase Equilibria. Ther-
modynamic Background and Practical Tools; John Wiley and Sons, Inc.:
New York, 1992.
Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 65
(53) Kooijman, H.; Taylor, R. The ChemSep Book; Library books on
demand: Norderstedt, 2000.
(54) Colombo, F.; Cori, L.; Dalioro, L.; Delogu, P. Equilibrium Constant
for the Methyl tert-Butyl Ether Liquid Phase Synthesis by Use of UNIFAC.
Ind. Eng. Chem. Fundam. 1983, 22, 219.
(55) Rehfinger, A.; Hoffmann, U. Kinetics of Methyl Tertiary Butyl
Ether Liquid Phase Synthesis Catalyzed by Ion Exchange ResinsI. Intrinsic
Rate Expression in Liquid Phase Activities. Chem. Eng. Sci. 1990, 45, 1605.
(56) Zhang, T.; Datta, R. Integral Analysis of Methyl tert-Butyl Ether
Synthesis Kinetics. Ind. Eng. Chem. Res. 1995, 34, 730.
(57) Izquierdo, J.; Cunill, F.; Vila, M.; Tejero, J.; Iborra, M. Equilibrium
Constants for Methyl tert-Butyl Ether Liquid-phase Synthesis. J. Chem.
Eng. Data 1992, 37, 339.
(58) Thiel, C.; Sundmacher, K.; Hoffmann, U. Residue Curve Maps for
Heterogeneously Catalysed Reactive Distillation of Fuel Ethers MTBE and
TAME. Chem. Eng. Sci. 2002, 52, 993.
(59) Seader, J. D.; Henley, E. J. Separation Process Principles; John
Wiley and Sons, Inc.: New York, 1998.
(60) Bansal, V.; Perkins, J.; Pistikopoulos, E.; Ross, R.; van Schijndel,
J. Simultaneous Design and Control Optimisation under Uncertainty.
Comput. Chem. Eng. 2000, 24, 261.
(61) Process Systems Enterprise. gPROMS AdVanced User Guide;
Process Systems Enterprise: London, 1999.
(62) Lim, Y. I.; Floquer, P.; Joulia, X.; Kim, S. D. Multiobjective
Optimization in Terms of Economics and Potential Environment Impact
for Process Design and Analysis in a Chemical Process Simulator. Ind.
Eng. Chem. Res. 1999, 38, 4729.
(63) Jimenez, E.; Salamon, P.; Rivero, R.; Rendon, C.; Hoffman, K.;
Schaller, M.; Andresen, B. Optimization of a Diabatic Column with
Sequential Heat Exchangers. Ind. Eng. Chem. Res. 2004, 43, 7566.
(64) Larsson, M.; Wang, C.; Dahl, J. Development of a Method for
Analysing Energy, Environmental and Economic Efficiency for an Integrated
Steel Plant. Appl. Therm. Eng. 2006, 26, 1353.
(65) Tsatsaronis, G. Definition and Nomenclature in Exergy Analysis
and Exergoeconomics. Energy 2007, 32, 249.
(66) Almeida-Rivera, C. P.; Swinkels, P. L. J.; Grievink, J. Economics
and Exergy Efficiency in the Conceptual Design of Reactive Distillation
Processes. AIChE Symp. Ser. 2004, 2004, 237.
(67) Huang, K.; Iwakabe, K.; Nakaiwa, M.; Tsutsumi, A. Towards
Further Internal Heat Integration of Reactive Distillation Columns. Part I:
The Design Principle. Chem. Eng. Sci. 2005, 60, 4901.
(68) Nummedal, L.; Kjelstrup, S. Equipartition of Forces as a Lower
Bound on the Entropy Production in Heat Exchange. Int. J. Heat Mass
Transfer 2001, 44, 2827.
(69) Sauar, E.; Siragusa, G.; Andresen, B. Equal Thermodynamic
Distance and Equipartition of Forces Principles Applied to Binary Distil-
lation. J. Phys. Chem. A 2001, 105, 2312.
(70) Georgiadis, M.; Schenk, M.; Pistikopoulos, E.; Gani, R. The
Interactions of Design, Control and Operability in Reactive Distillation
Systems. Comput. Chem. Eng. 2002, 26, 735.
(71) Wang, S. J.; Wong, D. S. H.; Lee, E. K. Effect of Interaction
Multiplicity on Control System Design for a MTBE Reactive Distillation
Column. J. Proc. Control 2003, 13, 503.
(72) Sneesby, M. G.; Tade, M. O.; Smith, T. N. Steady-state Transitions
in the Reactive Distillation of MTBE. Comput. Chem. Eng. 1998, 22, 879.
(73) Hasse, R. Thermodynamics of IrreVersible Processes; Addison-
Wesley Publishing Co.: Darmstadt, Germany, 1969.
(74) Ung, S.; Doherty, M. Vapor-Liquid Phase Equilibrium in Systems
with Multiple Chemical Reactors. Chem. Eng. Sci. 1995, 50, 23.
ReceiVed for reView November 28, 2006
ReVised manuscript receiVed September 10, 2007
Accepted September 11, 2007
IE061521J