(16b) Process Synthesis, Design and Intensification. an Integrated Approach

Process Synthesis,  Design and Intensification. An
integrated approach

Deenesh K. Babi¹,
John M. Woodley², Rafiqul Gani¹,
Jeffrey J. Siirola³

dkbabi@kt.dtu.dk, jw@kt.dtu.dk, rag@kt.dtu.dk, jjsiirola@gmail.com

¹ CAPEC-Department of Chemical and
Biochemical Engineering, Technical University of Denmark (DTU), DK-2800 Kongens Lyngby, Denmark

²PROCESS-Department of Chemical and
Biochemical Engineering, Technical University of Denmark (DTU), DK-2800 Kongens Lyngby, Denmark

³Department of Chemical Engineering,
Carnegie Mellon University, Pittsburgh, PA 15213-3890 USA

Abstract

Process design has four
main objectives: process synthesis or generation of process alternatives;
process analysis; process evaluation; and the selection of the best process to
achieve a desired goal [1]. According to Westerberg [2] the overall improvement
of a process, new or existing, may be achieved by considering  the original design as the base case,
performing flowsheet optimization to further improve
the design and through analysis of the real time
process operation. Possible improvements are identified and are generally,
where economically feasible, implemented during plant retrofits. Process intensification (PI) is a means by
which processes, whether conceptual or existing, can be designed or redesigned  to achieve
a more efficient and sustainable process through the
improvement of key process parameters, for example energy efficiency and waste
generation. PI is defined as the improvement of a process at the phenomena
level which ultimately has an impact at the higher levels of a process, for
example the functional and unit operations levels. More specifically it is the enhancement
achieved through the integration of unit operations, functions and/or phenomena.
Therefore incentives exist for the inclusion of PI within the overall concept
of process design and in particular as part of the process synthesis and design
methodology.

With the inclusion of PI into process design, not
only can a process be synthesised, analysed and evaluated but it can also be
further analysed for the potential use of intensified equipment within that
process, whether novel or mature. In order for this paradigm shift to occur the
intensified process should be better than the original design. An example of PI
was achieved by Eastman Chemical [3] which in 1983 intensified a process for
the manufacture of methyl acetate by replacing with one single reactive-extractive
distillation column a multi-step process which had consisted of a reactor, extractor,
decanter, several normal and azeotropic distillation
columns, and two mass separation agents.

For achieving PI via a systematic approach different
methods exist, for example, the task-based means-ends analysis developed by Siirola [1]. However, there is a need for not only a more
systematic, efficient and flexible PI method covering a wider range of
applications, but also for one that is able to find truly innovative and
predictive solutions, using not only the knowledge of existing methodologies at
the unit operations level but also at a lower phenomena
level, where new unit operations can be designed.

The objective of this work is twofold:  (1) The further development of a phenomena
based PI synthesis and design algorithm (PBS algorithm) and (2) To incorporate
the PBS algorithm as part of an overall process synthesis/design framework that
includes both task-based means-ends analysis and thermodynamic insights [4].  Within the overall synthesis/design framework
this PBS algorithm is employed to generate, screen and analyse many possible
alternatives. A first version of the PBS algorithm has been developed and
consists of a six steps.

The starting point of the PBS algorithm is the base
case design, which can either be the final design based on the current state of
the art of process design or the design of an existing process. In Step 1 the
problem is defined together with the objective function. In Step 2 a decomposition
approach is applied where the process is represented in terms of tasks and
process phenomena respectively and the process is then analysed using
thermodynamic insights for example analysis of pure component and mixture
properties. In Step 3 the limitations/bottlenecks (LBs) of the process can then
be identified together with the desirable and accompanying phenomena to
overcome these LBs. In Step 4 phenomena are connected
to form simultaneous phenomena building blocks (SPBs) which are screened for
the most feasible connections using for example connectivity rules. An example
of a connectivity rule is heating and cooling cannot exist within the same
SPBs. The SPB's are then connected to form operations which are then connected
to form flowsheets. In Step 5 these flowsheets are first screened using for example logical
constraints and performance metrics. In Step 6 the feasible flowhseets
from Step 5 are optimized with respect to the objective function defined in
Step 1 with the end result being the identification of the best intensified
process candidate. The emerging flowhseet consists of
either novel or existing equipment. Through the combined synthesis and design
framework, the results of the means-ends analysis and thermodynamic insights
methodologies are used in defining the problem (Step 1), in decomposing the
problem (Step 2), and in evaluating the results from Step 4.

In the proposed
presentation the overall process synthesis and design framework together with
its supporting algorithms and tools will be illustrated by a re-examination of
the methyl acetate process. It will be shown that the framework is able to
generate a wider range of solutions, including those that have been reported
earlier.

References:

[1] J. J. Siirola. Strategic process synthesis: Advances
in the hierarchical approach. Comput. Chem. Eng. 1996, Supplement 2 (20) S1637-S1643

[2] A. W. Westerberg. A retrospective on
design and process synthesis. Comput. Chem. Eng. 2004 (28) 447?458

[3]
V. H. Agreda, L. R. Partin
& W. H. Heise . High-purity methyl acetate via reactive
distillation. Chem. Eng. Prog. 1986 (2) 40?46

[4] C.
Jaksland, R. Gani, K. M. Lien. Separation process design and synthesis based on
thermodynamic insights. Chem Eng Sci, 1995 (50) 511-530