The Nature of Chemical Process Design and Integration - Chemical Engineering Chemical Engineering Notes | EduRev

Chemical Engineering : The Nature of Chemical Process Design and Integration - Chemical Engineering Chemical Engineering Notes | EduRev

 Page 1


1 The Nature of Chemical Process Design and Integration
1.1 CHEMICAL PRODUCTS
Chemical products are essential to modern living standards.
Almost all aspects of everyday life are supported by
chemical products in one way or another. Yet, society tends
to take these products for granted, even though a high
quality of life fundamentally depends on them.
When considering the design of processes for the
manufacture of chemical products, the market into which
they are being sold fundamentally in?uences the objectives
and priorities in the design. Chemical products can be
divided into three broad classes:
1. Commodity or bulk chemicals: These are produced in
large volumes and purchased on the basis of chemical
composition, purity and price. Examples are sulfuric
acid, nitrogen, oxygen, ethylene and chlorine.
2. Fine chemicals: These are produced in small volumes
and purchased on the basis of chemical composition,
purity and price. Examples are chloropropylene oxide
(used for the manufacture of epoxy resins, ion-exchange
resins and other products), dimethyl formamide (used,
for example, as a solvent, reaction medium and interme-
diate in the manufacture of pharmaceuticals), n-butyric
acid (used in beverages, ?avorings, fragrances and other
products) and barium titanate powder (used for the man-
ufacture of electronic capacitors).
3. Specialty or effect or functional chemicals:These are
purchased because of their effect (or function), rather
than their chemical composition. Examples are pharma-
ceuticals, pesticides, dyestuffs, perfumes and ?avorings.
Because commodity and ?ne chemicals tend to be pur-
chased on the basis of their chemical composition alone,
they are undifferentiated. For example, there is nothing to
choose between 99.9% benzene made by one manufacturer
and that made by another manufacturer, other than price
and delivery issues. On the other hand, specialty chemicals
tend to be purchased on the basis of their effect or function
and are therefore differentiated. For example, competitive
pharmaceutical products are differentiated according to the
ef?cacy of the product, rather than chemical composition.
An adhesive is purchased on the basis of its ability to stick
things together, rather than its chemical composition and
so on.
Chemical Process Design and Integration R. Smith
? 2005 John Wiley & Sons, Ltd ISBNs: 0-471-48680-9 (HB); 0-471-48681-7 (PB)
However, undifferentiated and differentiated should be
thought of as relative terms rather than absolute terms for
chemical products. In practice, chemicals do not tend to
be completely undifferentiated or completely differentiated.
Commodityand?nechemicalproductsmighthaveimpurity
speci?cations as well as purity speci?cations. Traces of
impurities can, in some cases, give some differentiation
between different manufacturers of commodity and ?ne
chemicals. For example, 99.9% acrylic acid might be
considered to be an undifferentiated product. However,
traces of impurities, at concentrations of a few parts per
million, can interfere with some of the reactions in which
it is used and can have important implications for some
of its uses. Such impurities might differ between different
manufacturing processes. Not all specialty products are
differentiated. For example, pharmaceutical products like
aspirin (acetylsalicylic acid) are undifferentiated. Different
manufacturers can produce aspirin and there is nothing to
choose between these products, other than the price and
differentiation created through marketing of the product.
Scale of production also differs between the three classes
of chemical products. Fine and specialty chemicals tend
to be produced in volumes less than 1000 t·y
-1
.Onthe
other hand, commodity chemicals tend to be produced in
much larger volumes than this. However, the distinction
is again not so clear. Polymers are differentiated products
because they are purchased on the basis of their mechanical
properties, but can be produced in quantities signi?cantly
higher than 1000 t·y
-1
.
When a new chemical product is ?rst developed, it
can often be protected by a patent in the early years of
commercial exploitation. For a product to be eligible to
be patented, it must be novel, useful and unobvious. If
patent protection can be obtained, this effectively gives
the producer a monopoly for commercial exploitation of
the product until the patent expires. Patent protection lasts
for 20 years from the ?ling date of the patent. Once the
patent expires, competitors can join in and manufacture the
product. If competitors cannot wait until the patent expires,
then alternative competing products must be developed.
Another way to protect a competitive edge for a new
product is to protect it by secrecy. The formula for Coca-
Cola has been kept a secret for over 100 years. Potentially,
there is no time limit on such protection. However, for
the protection through secrecy to be viable, competitors
must not be able to reproduce the product from chemical
analysis. This is likely to be the caseonly for certain classes
of specialty and food products for which the properties of
Page 2


1 The Nature of Chemical Process Design and Integration
1.1 CHEMICAL PRODUCTS
Chemical products are essential to modern living standards.
Almost all aspects of everyday life are supported by
chemical products in one way or another. Yet, society tends
to take these products for granted, even though a high
quality of life fundamentally depends on them.
When considering the design of processes for the
manufacture of chemical products, the market into which
they are being sold fundamentally in?uences the objectives
and priorities in the design. Chemical products can be
divided into three broad classes:
1. Commodity or bulk chemicals: These are produced in
large volumes and purchased on the basis of chemical
composition, purity and price. Examples are sulfuric
acid, nitrogen, oxygen, ethylene and chlorine.
2. Fine chemicals: These are produced in small volumes
and purchased on the basis of chemical composition,
purity and price. Examples are chloropropylene oxide
(used for the manufacture of epoxy resins, ion-exchange
resins and other products), dimethyl formamide (used,
for example, as a solvent, reaction medium and interme-
diate in the manufacture of pharmaceuticals), n-butyric
acid (used in beverages, ?avorings, fragrances and other
products) and barium titanate powder (used for the man-
ufacture of electronic capacitors).
3. Specialty or effect or functional chemicals:These are
purchased because of their effect (or function), rather
than their chemical composition. Examples are pharma-
ceuticals, pesticides, dyestuffs, perfumes and ?avorings.
Because commodity and ?ne chemicals tend to be pur-
chased on the basis of their chemical composition alone,
they are undifferentiated. For example, there is nothing to
choose between 99.9% benzene made by one manufacturer
and that made by another manufacturer, other than price
and delivery issues. On the other hand, specialty chemicals
tend to be purchased on the basis of their effect or function
and are therefore differentiated. For example, competitive
pharmaceutical products are differentiated according to the
ef?cacy of the product, rather than chemical composition.
An adhesive is purchased on the basis of its ability to stick
things together, rather than its chemical composition and
so on.
Chemical Process Design and Integration R. Smith
? 2005 John Wiley & Sons, Ltd ISBNs: 0-471-48680-9 (HB); 0-471-48681-7 (PB)
However, undifferentiated and differentiated should be
thought of as relative terms rather than absolute terms for
chemical products. In practice, chemicals do not tend to
be completely undifferentiated or completely differentiated.
Commodityand?nechemicalproductsmighthaveimpurity
speci?cations as well as purity speci?cations. Traces of
impurities can, in some cases, give some differentiation
between different manufacturers of commodity and ?ne
chemicals. For example, 99.9% acrylic acid might be
considered to be an undifferentiated product. However,
traces of impurities, at concentrations of a few parts per
million, can interfere with some of the reactions in which
it is used and can have important implications for some
of its uses. Such impurities might differ between different
manufacturing processes. Not all specialty products are
differentiated. For example, pharmaceutical products like
aspirin (acetylsalicylic acid) are undifferentiated. Different
manufacturers can produce aspirin and there is nothing to
choose between these products, other than the price and
differentiation created through marketing of the product.
Scale of production also differs between the three classes
of chemical products. Fine and specialty chemicals tend
to be produced in volumes less than 1000 t·y
-1
.Onthe
other hand, commodity chemicals tend to be produced in
much larger volumes than this. However, the distinction
is again not so clear. Polymers are differentiated products
because they are purchased on the basis of their mechanical
properties, but can be produced in quantities signi?cantly
higher than 1000 t·y
-1
.
When a new chemical product is ?rst developed, it
can often be protected by a patent in the early years of
commercial exploitation. For a product to be eligible to
be patented, it must be novel, useful and unobvious. If
patent protection can be obtained, this effectively gives
the producer a monopoly for commercial exploitation of
the product until the patent expires. Patent protection lasts
for 20 years from the ?ling date of the patent. Once the
patent expires, competitors can join in and manufacture the
product. If competitors cannot wait until the patent expires,
then alternative competing products must be developed.
Another way to protect a competitive edge for a new
product is to protect it by secrecy. The formula for Coca-
Cola has been kept a secret for over 100 years. Potentially,
there is no time limit on such protection. However, for
the protection through secrecy to be viable, competitors
must not be able to reproduce the product from chemical
analysis. This is likely to be the caseonly for certain classes
of specialty and food products for which the properties of
2 The Nature of Chemical Process Design and Integration
the product depend on both the chemical composition and
the method of manufacture.
Figure 1.1 illustrates different product life cycles
1,2
.The
general trend is that when a new product is introduced
into the market, the sales grow slowly until the market
is established and then more rapidly once the market is
established. If there is patent protection, then competitors
will not be able to exploit the same product commercially
until the patent expires, when competitors can produce the
same product and take market share. It is expected that
competitive products will cause sales to diminish later in
the product life cycle until sales become so low that a
company would be expected to withdraw from the market.
In Figure 1.1, Product A appears to be a poor product that
has a short life with low sales volume. It might be that it
cannot compete well with other competitive products, and
alternative products quickly force the company out of that
business. However, a low sales volume is not the main
criterion to withdraw from the market. It might be that
a product with low volume ?nds a market niche and can
be sold for a high value. On the other hand, if it were
competing with other products with similar functions in
the same market sector, which keeps both the sale price
and volume low, then it would seem wise to withdraw
from the market. Product B in Figure 1.1 appears to be
a better product, showing a longer life cycle and higher
sales volume. This has patent protection but sales decrease
rapidly after patent protection is lost, leading to loss of
market through competition. Product C in Figure 1.1 is
a still better product. This shows high sales volume with
the life of the product extended through reformulation of
the product
1
. Finally, Product D in Figure 1.1 shows a
product life cycle that is typical of commodity chemicals.
Commodity chemicals tend not to exhibit the same kind
of life cycles as ?ne and specialty chemicals. In the early
years of the commercial exploitation, the sales volume
grows rapidly to a high volume, but then does not decline
and enters a mature period of slow growth, or, in some
exceptional cases, slow decline. This is because commodity
chemicals tend to havea diverse rangeof uses. Even though
competition might take away some end uses, new end uses
are introduced, leading to an extended life cycle.
The different classes of chemical products will have
very different added value (the difference between the
selling price of the product and the purchase cost of
raw materials). Commodity chemicals tend to have low
added value, whereas ?ne and specialty chemicals tend to
have high added value. Commodity chemicals tend to be
produced in large volumes with low added value, while
?ne and specialty chemicals tend to be produced in small
volumes with high added value.
Because of this, when designing a process for a
commodity chemical, it is usually important to keep
operating costs as low as possible. The capital cost of the
process will tend to be high relative to a process for ?ne or
specialty chemicals because of the scale of production.
When designing a process for specialty chemicals,
priority tends to be given to the product, rather than to
the process. This is because the unique function of the
product must be protected. The process is likely to be
small scale and operating costs tend to be less important
than with commodity chemical processes. The capital
cost of the process will be low relative to commodity
chemical processes because of the scale. The time to
Product
Sales
(t • y
-1
)
Product D 
Product C 
Product B 
Product A 
Patent Expiry
Product
Reformulation
Time
(y )
Figure 1.1 Product life cycles. (Adapted from Sharratt PN, 1997, Handbook of Batch Process Design, Blackie Academic and
Professional by permission).
Page 3


1 The Nature of Chemical Process Design and Integration
1.1 CHEMICAL PRODUCTS
Chemical products are essential to modern living standards.
Almost all aspects of everyday life are supported by
chemical products in one way or another. Yet, society tends
to take these products for granted, even though a high
quality of life fundamentally depends on them.
When considering the design of processes for the
manufacture of chemical products, the market into which
they are being sold fundamentally in?uences the objectives
and priorities in the design. Chemical products can be
divided into three broad classes:
1. Commodity or bulk chemicals: These are produced in
large volumes and purchased on the basis of chemical
composition, purity and price. Examples are sulfuric
acid, nitrogen, oxygen, ethylene and chlorine.
2. Fine chemicals: These are produced in small volumes
and purchased on the basis of chemical composition,
purity and price. Examples are chloropropylene oxide
(used for the manufacture of epoxy resins, ion-exchange
resins and other products), dimethyl formamide (used,
for example, as a solvent, reaction medium and interme-
diate in the manufacture of pharmaceuticals), n-butyric
acid (used in beverages, ?avorings, fragrances and other
products) and barium titanate powder (used for the man-
ufacture of electronic capacitors).
3. Specialty or effect or functional chemicals:These are
purchased because of their effect (or function), rather
than their chemical composition. Examples are pharma-
ceuticals, pesticides, dyestuffs, perfumes and ?avorings.
Because commodity and ?ne chemicals tend to be pur-
chased on the basis of their chemical composition alone,
they are undifferentiated. For example, there is nothing to
choose between 99.9% benzene made by one manufacturer
and that made by another manufacturer, other than price
and delivery issues. On the other hand, specialty chemicals
tend to be purchased on the basis of their effect or function
and are therefore differentiated. For example, competitive
pharmaceutical products are differentiated according to the
ef?cacy of the product, rather than chemical composition.
An adhesive is purchased on the basis of its ability to stick
things together, rather than its chemical composition and
so on.
Chemical Process Design and Integration R. Smith
? 2005 John Wiley & Sons, Ltd ISBNs: 0-471-48680-9 (HB); 0-471-48681-7 (PB)
However, undifferentiated and differentiated should be
thought of as relative terms rather than absolute terms for
chemical products. In practice, chemicals do not tend to
be completely undifferentiated or completely differentiated.
Commodityand?nechemicalproductsmighthaveimpurity
speci?cations as well as purity speci?cations. Traces of
impurities can, in some cases, give some differentiation
between different manufacturers of commodity and ?ne
chemicals. For example, 99.9% acrylic acid might be
considered to be an undifferentiated product. However,
traces of impurities, at concentrations of a few parts per
million, can interfere with some of the reactions in which
it is used and can have important implications for some
of its uses. Such impurities might differ between different
manufacturing processes. Not all specialty products are
differentiated. For example, pharmaceutical products like
aspirin (acetylsalicylic acid) are undifferentiated. Different
manufacturers can produce aspirin and there is nothing to
choose between these products, other than the price and
differentiation created through marketing of the product.
Scale of production also differs between the three classes
of chemical products. Fine and specialty chemicals tend
to be produced in volumes less than 1000 t·y
-1
.Onthe
other hand, commodity chemicals tend to be produced in
much larger volumes than this. However, the distinction
is again not so clear. Polymers are differentiated products
because they are purchased on the basis of their mechanical
properties, but can be produced in quantities signi?cantly
higher than 1000 t·y
-1
.
When a new chemical product is ?rst developed, it
can often be protected by a patent in the early years of
commercial exploitation. For a product to be eligible to
be patented, it must be novel, useful and unobvious. If
patent protection can be obtained, this effectively gives
the producer a monopoly for commercial exploitation of
the product until the patent expires. Patent protection lasts
for 20 years from the ?ling date of the patent. Once the
patent expires, competitors can join in and manufacture the
product. If competitors cannot wait until the patent expires,
then alternative competing products must be developed.
Another way to protect a competitive edge for a new
product is to protect it by secrecy. The formula for Coca-
Cola has been kept a secret for over 100 years. Potentially,
there is no time limit on such protection. However, for
the protection through secrecy to be viable, competitors
must not be able to reproduce the product from chemical
analysis. This is likely to be the caseonly for certain classes
of specialty and food products for which the properties of
2 The Nature of Chemical Process Design and Integration
the product depend on both the chemical composition and
the method of manufacture.
Figure 1.1 illustrates different product life cycles
1,2
.The
general trend is that when a new product is introduced
into the market, the sales grow slowly until the market
is established and then more rapidly once the market is
established. If there is patent protection, then competitors
will not be able to exploit the same product commercially
until the patent expires, when competitors can produce the
same product and take market share. It is expected that
competitive products will cause sales to diminish later in
the product life cycle until sales become so low that a
company would be expected to withdraw from the market.
In Figure 1.1, Product A appears to be a poor product that
has a short life with low sales volume. It might be that it
cannot compete well with other competitive products, and
alternative products quickly force the company out of that
business. However, a low sales volume is not the main
criterion to withdraw from the market. It might be that
a product with low volume ?nds a market niche and can
be sold for a high value. On the other hand, if it were
competing with other products with similar functions in
the same market sector, which keeps both the sale price
and volume low, then it would seem wise to withdraw
from the market. Product B in Figure 1.1 appears to be
a better product, showing a longer life cycle and higher
sales volume. This has patent protection but sales decrease
rapidly after patent protection is lost, leading to loss of
market through competition. Product C in Figure 1.1 is
a still better product. This shows high sales volume with
the life of the product extended through reformulation of
the product
1
. Finally, Product D in Figure 1.1 shows a
product life cycle that is typical of commodity chemicals.
Commodity chemicals tend not to exhibit the same kind
of life cycles as ?ne and specialty chemicals. In the early
years of the commercial exploitation, the sales volume
grows rapidly to a high volume, but then does not decline
and enters a mature period of slow growth, or, in some
exceptional cases, slow decline. This is because commodity
chemicals tend to havea diverse rangeof uses. Even though
competition might take away some end uses, new end uses
are introduced, leading to an extended life cycle.
The different classes of chemical products will have
very different added value (the difference between the
selling price of the product and the purchase cost of
raw materials). Commodity chemicals tend to have low
added value, whereas ?ne and specialty chemicals tend to
have high added value. Commodity chemicals tend to be
produced in large volumes with low added value, while
?ne and specialty chemicals tend to be produced in small
volumes with high added value.
Because of this, when designing a process for a
commodity chemical, it is usually important to keep
operating costs as low as possible. The capital cost of the
process will tend to be high relative to a process for ?ne or
specialty chemicals because of the scale of production.
When designing a process for specialty chemicals,
priority tends to be given to the product, rather than to
the process. This is because the unique function of the
product must be protected. The process is likely to be
small scale and operating costs tend to be less important
than with commodity chemical processes. The capital
cost of the process will be low relative to commodity
chemical processes because of the scale. The time to
Product
Sales
(t • y
-1
)
Product D 
Product C 
Product B 
Product A 
Patent Expiry
Product
Reformulation
Time
(y )
Figure 1.1 Product life cycles. (Adapted from Sharratt PN, 1997, Handbook of Batch Process Design, Blackie Academic and
Professional by permission).
Formulation of the Design Problem 3
market the product is also likely to be important with
specialty chemicals, especially if there is patent protection.
If this is the case, then anything that shortens the time
from basic research, through product testing, pilot plant
studies, process design, construction of the plant to product
manufacture will have an important in?uence on the overall
project pro?tability.
All this means that the priorities in process design are
likely to differ signi?cantly, depending on whether a pro-
cess is being designed for the manufacture of a commodity,
?ne or specialty chemical. In commodity chemicals, thereis
likely to berelatively little productinnovation, butintensive
process innovation. Also, equipment will be designed for a
speci?c process step. On the other hand, the manufacture
of ?ne and specialty chemicals might involve:
• selling into a market with low volume,
• short product life cycle,
• a demand for a short time to market, and therefore, less
time is available for process development, with product
and process development proceeding simultaneously.
Because of this, the manufacture of ?ne and specialty
chemicals is often carried out in multipurpose equipment,
perhaps with different chemicals being manufactured in
the same equipment at different times during the year.
The life of the equipment might greatly exceed the life
of the product.
The development of pharmaceutical products is such that
high-quality products must be manufactured during the
development of the process to allow safety and clinical
studies to be carried out before full-scale production.
Pharmaceutical production represents an extreme case
of process design in which the regulatory framework
controlling production makes it dif?cult to make process
changes, even during the development stage. Even if
signi?cant improvements to processes for pharmaceuticals
can be suggested, it might not be feasible to implement
them, as such changes might prevent or delay the process
from being licensed for production.
1.2 FORMULATION OF THE DESIGN
PROBLEM
Before a process design can be started, the design prob-
lem must be formulated. Formulation of the design problem
requires a product speci?cation. If a well-de?ned chemical
product is to be manufactured, then the speci?cation of the
productmightappearstraightforward(e.g.apurifyspeci?ca-
tion). However,if aspecialty productis to bemanufactured,
it is the functional properties that are important, rather than
the chemical properties, and this might require a product
design stageinordertospecifytheproduct
3
.Theinitialstate-
mentofthedesignproblemisoftenillde?ned.Forexample,
the design team could be asked to expand the production
capacityof anexisting plantthatproducesa chemicalthatis
a precursor to a polymer product, which is also produced by
the company. This results from an increase in the demand
for the polymer product and the plant producing the precur-
sor currently being operated at its maximum capacity. The
designer might well be given a speci?cation for the expan-
sion. For example, the marketing department might assess
that the market could be expanded by 30% over a two-year
period, which would justify a 30% expansion in the process
for the precursor. However, the 30% projection can easily
be wrong. The economic environment can change, leading
to the projected increase being either too large or too small.
It might also be possible to sell the polymer precursor in the
market to other manufacturers of the polymer and justify an
expansionevenlargerthan30%.Ifthepolymerprecursorcan
besoldinthemarketplace,isthecurrentpurityspeci?cation
of the company suitable for the marketplace? Perhaps the
marketplace demands a higher purity than what is currently
the company speci?cation. Perhaps the current speci?cation
is acceptable,but if the speci?cation could be improved, the
product could be sold for a higher value and/or at a greater
volume.Anoption might betonotexpandtheproductionof
thepolymerprecursorto30%,butinsteadtopurchaseitfrom
the market. If it is purchased from the market, is it likely to
be up to the company speci?cations, or will it need some
puri?cation before it is suitable for the company’s polymer
process? How reliable will the market source be? All these
uncertainties are related more to market supply and demand
issues than to speci?c process design issues.
Closer examination of the current process design might
lead to the conclusion that the capacity can be expanded
by 10% with a very modest capital investment. A
further increase to 20% would require a signi?cant capital
investment, but an expansion to 30% would require an
extremely large capital investment. This opens up further
options. Should the plant be expanded by 10% and a
market source identi?ed for the balance? Should the plant
be expanded to 20% similarly? If a real expansion in the
market place is anticipated and expansion to 30% would
be very expensive, why not be more aggressive and instead
of expanding the existing process, build an entirely new
process? If a new process is to be built, then what should
be the process technology? New process technology might
have been developed since the original plant was built
that enables the same product to be manufactured at a
much lower cost. If a new process is to be built, where
should it be built? It might make more sense to build
it in another country that would allow lower operating
costs, and the product could be shipped back to be fed
to the existing polymer process. At the same time, this
might stimulate the development of new markets in other
countries, in which case, what should be the capacity of the
new plant?
Fromallof these options, thedesign teammust formulate
a number of plausible design options. Thus, from the initial
Page 4


1 The Nature of Chemical Process Design and Integration
1.1 CHEMICAL PRODUCTS
Chemical products are essential to modern living standards.
Almost all aspects of everyday life are supported by
chemical products in one way or another. Yet, society tends
to take these products for granted, even though a high
quality of life fundamentally depends on them.
When considering the design of processes for the
manufacture of chemical products, the market into which
they are being sold fundamentally in?uences the objectives
and priorities in the design. Chemical products can be
divided into three broad classes:
1. Commodity or bulk chemicals: These are produced in
large volumes and purchased on the basis of chemical
composition, purity and price. Examples are sulfuric
acid, nitrogen, oxygen, ethylene and chlorine.
2. Fine chemicals: These are produced in small volumes
and purchased on the basis of chemical composition,
purity and price. Examples are chloropropylene oxide
(used for the manufacture of epoxy resins, ion-exchange
resins and other products), dimethyl formamide (used,
for example, as a solvent, reaction medium and interme-
diate in the manufacture of pharmaceuticals), n-butyric
acid (used in beverages, ?avorings, fragrances and other
products) and barium titanate powder (used for the man-
ufacture of electronic capacitors).
3. Specialty or effect or functional chemicals:These are
purchased because of their effect (or function), rather
than their chemical composition. Examples are pharma-
ceuticals, pesticides, dyestuffs, perfumes and ?avorings.
Because commodity and ?ne chemicals tend to be pur-
chased on the basis of their chemical composition alone,
they are undifferentiated. For example, there is nothing to
choose between 99.9% benzene made by one manufacturer
and that made by another manufacturer, other than price
and delivery issues. On the other hand, specialty chemicals
tend to be purchased on the basis of their effect or function
and are therefore differentiated. For example, competitive
pharmaceutical products are differentiated according to the
ef?cacy of the product, rather than chemical composition.
An adhesive is purchased on the basis of its ability to stick
things together, rather than its chemical composition and
so on.
Chemical Process Design and Integration R. Smith
? 2005 John Wiley & Sons, Ltd ISBNs: 0-471-48680-9 (HB); 0-471-48681-7 (PB)
However, undifferentiated and differentiated should be
thought of as relative terms rather than absolute terms for
chemical products. In practice, chemicals do not tend to
be completely undifferentiated or completely differentiated.
Commodityand?nechemicalproductsmighthaveimpurity
speci?cations as well as purity speci?cations. Traces of
impurities can, in some cases, give some differentiation
between different manufacturers of commodity and ?ne
chemicals. For example, 99.9% acrylic acid might be
considered to be an undifferentiated product. However,
traces of impurities, at concentrations of a few parts per
million, can interfere with some of the reactions in which
it is used and can have important implications for some
of its uses. Such impurities might differ between different
manufacturing processes. Not all specialty products are
differentiated. For example, pharmaceutical products like
aspirin (acetylsalicylic acid) are undifferentiated. Different
manufacturers can produce aspirin and there is nothing to
choose between these products, other than the price and
differentiation created through marketing of the product.
Scale of production also differs between the three classes
of chemical products. Fine and specialty chemicals tend
to be produced in volumes less than 1000 t·y
-1
.Onthe
other hand, commodity chemicals tend to be produced in
much larger volumes than this. However, the distinction
is again not so clear. Polymers are differentiated products
because they are purchased on the basis of their mechanical
properties, but can be produced in quantities signi?cantly
higher than 1000 t·y
-1
.
When a new chemical product is ?rst developed, it
can often be protected by a patent in the early years of
commercial exploitation. For a product to be eligible to
be patented, it must be novel, useful and unobvious. If
patent protection can be obtained, this effectively gives
the producer a monopoly for commercial exploitation of
the product until the patent expires. Patent protection lasts
for 20 years from the ?ling date of the patent. Once the
patent expires, competitors can join in and manufacture the
product. If competitors cannot wait until the patent expires,
then alternative competing products must be developed.
Another way to protect a competitive edge for a new
product is to protect it by secrecy. The formula for Coca-
Cola has been kept a secret for over 100 years. Potentially,
there is no time limit on such protection. However, for
the protection through secrecy to be viable, competitors
must not be able to reproduce the product from chemical
analysis. This is likely to be the caseonly for certain classes
of specialty and food products for which the properties of
2 The Nature of Chemical Process Design and Integration
the product depend on both the chemical composition and
the method of manufacture.
Figure 1.1 illustrates different product life cycles
1,2
.The
general trend is that when a new product is introduced
into the market, the sales grow slowly until the market
is established and then more rapidly once the market is
established. If there is patent protection, then competitors
will not be able to exploit the same product commercially
until the patent expires, when competitors can produce the
same product and take market share. It is expected that
competitive products will cause sales to diminish later in
the product life cycle until sales become so low that a
company would be expected to withdraw from the market.
In Figure 1.1, Product A appears to be a poor product that
has a short life with low sales volume. It might be that it
cannot compete well with other competitive products, and
alternative products quickly force the company out of that
business. However, a low sales volume is not the main
criterion to withdraw from the market. It might be that
a product with low volume ?nds a market niche and can
be sold for a high value. On the other hand, if it were
competing with other products with similar functions in
the same market sector, which keeps both the sale price
and volume low, then it would seem wise to withdraw
from the market. Product B in Figure 1.1 appears to be
a better product, showing a longer life cycle and higher
sales volume. This has patent protection but sales decrease
rapidly after patent protection is lost, leading to loss of
market through competition. Product C in Figure 1.1 is
a still better product. This shows high sales volume with
the life of the product extended through reformulation of
the product
1
. Finally, Product D in Figure 1.1 shows a
product life cycle that is typical of commodity chemicals.
Commodity chemicals tend not to exhibit the same kind
of life cycles as ?ne and specialty chemicals. In the early
years of the commercial exploitation, the sales volume
grows rapidly to a high volume, but then does not decline
and enters a mature period of slow growth, or, in some
exceptional cases, slow decline. This is because commodity
chemicals tend to havea diverse rangeof uses. Even though
competition might take away some end uses, new end uses
are introduced, leading to an extended life cycle.
The different classes of chemical products will have
very different added value (the difference between the
selling price of the product and the purchase cost of
raw materials). Commodity chemicals tend to have low
added value, whereas ?ne and specialty chemicals tend to
have high added value. Commodity chemicals tend to be
produced in large volumes with low added value, while
?ne and specialty chemicals tend to be produced in small
volumes with high added value.
Because of this, when designing a process for a
commodity chemical, it is usually important to keep
operating costs as low as possible. The capital cost of the
process will tend to be high relative to a process for ?ne or
specialty chemicals because of the scale of production.
When designing a process for specialty chemicals,
priority tends to be given to the product, rather than to
the process. This is because the unique function of the
product must be protected. The process is likely to be
small scale and operating costs tend to be less important
than with commodity chemical processes. The capital
cost of the process will be low relative to commodity
chemical processes because of the scale. The time to
Product
Sales
(t • y
-1
)
Product D 
Product C 
Product B 
Product A 
Patent Expiry
Product
Reformulation
Time
(y )
Figure 1.1 Product life cycles. (Adapted from Sharratt PN, 1997, Handbook of Batch Process Design, Blackie Academic and
Professional by permission).
Formulation of the Design Problem 3
market the product is also likely to be important with
specialty chemicals, especially if there is patent protection.
If this is the case, then anything that shortens the time
from basic research, through product testing, pilot plant
studies, process design, construction of the plant to product
manufacture will have an important in?uence on the overall
project pro?tability.
All this means that the priorities in process design are
likely to differ signi?cantly, depending on whether a pro-
cess is being designed for the manufacture of a commodity,
?ne or specialty chemical. In commodity chemicals, thereis
likely to berelatively little productinnovation, butintensive
process innovation. Also, equipment will be designed for a
speci?c process step. On the other hand, the manufacture
of ?ne and specialty chemicals might involve:
• selling into a market with low volume,
• short product life cycle,
• a demand for a short time to market, and therefore, less
time is available for process development, with product
and process development proceeding simultaneously.
Because of this, the manufacture of ?ne and specialty
chemicals is often carried out in multipurpose equipment,
perhaps with different chemicals being manufactured in
the same equipment at different times during the year.
The life of the equipment might greatly exceed the life
of the product.
The development of pharmaceutical products is such that
high-quality products must be manufactured during the
development of the process to allow safety and clinical
studies to be carried out before full-scale production.
Pharmaceutical production represents an extreme case
of process design in which the regulatory framework
controlling production makes it dif?cult to make process
changes, even during the development stage. Even if
signi?cant improvements to processes for pharmaceuticals
can be suggested, it might not be feasible to implement
them, as such changes might prevent or delay the process
from being licensed for production.
1.2 FORMULATION OF THE DESIGN
PROBLEM
Before a process design can be started, the design prob-
lem must be formulated. Formulation of the design problem
requires a product speci?cation. If a well-de?ned chemical
product is to be manufactured, then the speci?cation of the
productmightappearstraightforward(e.g.apurifyspeci?ca-
tion). However,if aspecialty productis to bemanufactured,
it is the functional properties that are important, rather than
the chemical properties, and this might require a product
design stageinordertospecifytheproduct
3
.Theinitialstate-
mentofthedesignproblemisoftenillde?ned.Forexample,
the design team could be asked to expand the production
capacityof anexisting plantthatproducesa chemicalthatis
a precursor to a polymer product, which is also produced by
the company. This results from an increase in the demand
for the polymer product and the plant producing the precur-
sor currently being operated at its maximum capacity. The
designer might well be given a speci?cation for the expan-
sion. For example, the marketing department might assess
that the market could be expanded by 30% over a two-year
period, which would justify a 30% expansion in the process
for the precursor. However, the 30% projection can easily
be wrong. The economic environment can change, leading
to the projected increase being either too large or too small.
It might also be possible to sell the polymer precursor in the
market to other manufacturers of the polymer and justify an
expansionevenlargerthan30%.Ifthepolymerprecursorcan
besoldinthemarketplace,isthecurrentpurityspeci?cation
of the company suitable for the marketplace? Perhaps the
marketplace demands a higher purity than what is currently
the company speci?cation. Perhaps the current speci?cation
is acceptable,but if the speci?cation could be improved, the
product could be sold for a higher value and/or at a greater
volume.Anoption might betonotexpandtheproductionof
thepolymerprecursorto30%,butinsteadtopurchaseitfrom
the market. If it is purchased from the market, is it likely to
be up to the company speci?cations, or will it need some
puri?cation before it is suitable for the company’s polymer
process? How reliable will the market source be? All these
uncertainties are related more to market supply and demand
issues than to speci?c process design issues.
Closer examination of the current process design might
lead to the conclusion that the capacity can be expanded
by 10% with a very modest capital investment. A
further increase to 20% would require a signi?cant capital
investment, but an expansion to 30% would require an
extremely large capital investment. This opens up further
options. Should the plant be expanded by 10% and a
market source identi?ed for the balance? Should the plant
be expanded to 20% similarly? If a real expansion in the
market place is anticipated and expansion to 30% would
be very expensive, why not be more aggressive and instead
of expanding the existing process, build an entirely new
process? If a new process is to be built, then what should
be the process technology? New process technology might
have been developed since the original plant was built
that enables the same product to be manufactured at a
much lower cost. If a new process is to be built, where
should it be built? It might make more sense to build
it in another country that would allow lower operating
costs, and the product could be shipped back to be fed
to the existing polymer process. At the same time, this
might stimulate the development of new markets in other
countries, in which case, what should be the capacity of the
new plant?
Fromallof these options, thedesign teammust formulate
a number of plausible design options. Thus, from the initial
4 The Nature of Chemical Process Design and Integration
ill-de?ned problem, the design team must create a series of
very speci?c options and these should then be compared
on the basis of a common set of assumptions regarding, for
example, raw materials prices and product prices. Having
speci?ed an option, this gives the design team a well-
de?ned problem to which the methods of engineering and
economic analysis can be applied.
In examining a design option, the design team should
start out by examining the problem at the highest level,
in terms of its feasibility with the minimum of detail to
ensure the design option is worth progressing
4
.Istherea
large difference between the value of the product and the
cost of the raw materials? If the overall feasibility looks
attractive, then more detail can be added, the option re-
evaluated,furtherdetailadded,andsoon.Byproductsmight
play a particularly important role in the economics. It might
be that the current process produces some byproducts that
can be sold in small quantities to the market. But, as the
process is expanded, there might be market constraints for
the new scale of production. If the byproducts cannot be
sold, how does this affect the economics?
If the design option appears to be technically and eco-
nomically feasible, then additional detail can be considered.
Material and energy balances can be formulated to give
a better de?nition to the inner workings of the process
and a more detailed process design can be developed. The
design calculations for this will normally be solved to a
high level of precision. However, a high level of preci-
sion cannot usually be justi?ed in terms of the operation of
the plant after it has been built. The plant will almost never
workpreciselyatitsoriginaldesign?owrates,temperatures,
pressures and compositions. This might be because the raw
materials are slightly different than what is assumed in the
design. The physical properties assumed in the calculations
might have been erroneous in some way, or operation at the
original design conditions might create corrosion or foul-
ing problems, or perhaps the plant cannot be controlled
adequately at the original conditions, and so on, for a mul-
titude of other possible reasons. The instrumentation on
the plant will not be able to measure the ?owrates, tem-
peratures, pressures and compositions as accurately as the
calculations performed. High precision might be required
for certain speci?c parts of the design. For example, the
polymer precursor might need certain impurities to be very
tightly controlled, perhaps down to the level of parts per
million.Itmightbethatsomecontaminantinawastestream
might be exceptionally environmentally harmful and must
be extremely well de?ned in the design calculations.
Even though a high level of precision cannot be justi?ed
in many cases in terms of the plant operation, the design
calculations will normally be carried out to a reasonably
high level of precision. The value of precision in design
calculationsisthattheconsistencyofthecalculationscanbe
checkedtoallowerrorsorpoorassumptionstobeidenti?ed.
It also allows the design options to be compared on a valid
like-for-like basis.
Because of all the uncertainties in carrying out a design,
thespeci?cationsareoftenincreasedbeyondthoseindicated
by the design calculations and the plant is overdesigned,
or contingency is added, through the application of safety
factors to the design. For example, the designer might
calculate the number of distillation plates required for a
distillation separation using elaborate calculations to a high
degree of precision, only to add an arbitrary extra 10% to
the number of plates for contingency. This allows for the
feed to the unit not being exactly as speci?ed, errors in the
physical properties, upset conditions in the plant, control
requirements, and so on. If too little contingency is added,
the plant might not work. If too much contingency is added,
the plant will not only be unnecessarily expensive, but too
much overdesign might make the plant dif?cult to operate
and might lead to a less ef?cient plant. For example, the
designer might calculate the size of a heat exchanger and
then add in a large contingency and signi?cantly oversize
the heat exchanger. The lower ?uid velocities encountered
bytheoversizedheatexchangercancauseittohaveapoorer
performance and to foul up more readily than a smaller
heat exchanger. Thus, a balance must be made between
different risks.
Insummary,theoriginalproblemposedtoprocessdesign
teams is often ill-de?ned, even though it might appear to
be well de?ned in the original design speci?cation. The
designteammustthenformulateaseriesofplausibledesign
options to be screened by the methods of engineering and
economic analysis. These design options are formulated
into very speci?c design problems. Some design options
mightbeeliminatedearlybyhigh-levelargumentsorsimple
calculations. Others will require more detailed examination.
In this way, the design team turns the ill-de?ned problem
into a well-de?ned one for analysis. To allow for the many
unquanti?able uncertainties, overdesign is used. Too little
overdesign might lead to the plant not working. Too much
overdesign will lead to the plant becoming unnecessarily
expensive,andperhapsdif?culttooperateandlessef?cient.
A balance must be made between different risks.
Consider the basic features of the design of chemical
processes now.
1.3 CHEMICAL PROCESS DESIGN
AND INTEGRATION
In a chemical process, the transformation of raw materials
into desired chemical products usually cannot be achieved
in a single step. Instead, the overall transformation is bro-
ken down into a number of steps that provide intermediate
transformations.Thesearecarriedoutthroughreaction,sep-
aration, mixing, heating, cooling, pressure change, particle
size reduction or enlargement. Once individual steps have
been selected, they must be interconnected to carry out the
Page 5


1 The Nature of Chemical Process Design and Integration
1.1 CHEMICAL PRODUCTS
Chemical products are essential to modern living standards.
Almost all aspects of everyday life are supported by
chemical products in one way or another. Yet, society tends
to take these products for granted, even though a high
quality of life fundamentally depends on them.
When considering the design of processes for the
manufacture of chemical products, the market into which
they are being sold fundamentally in?uences the objectives
and priorities in the design. Chemical products can be
divided into three broad classes:
1. Commodity or bulk chemicals: These are produced in
large volumes and purchased on the basis of chemical
composition, purity and price. Examples are sulfuric
acid, nitrogen, oxygen, ethylene and chlorine.
2. Fine chemicals: These are produced in small volumes
and purchased on the basis of chemical composition,
purity and price. Examples are chloropropylene oxide
(used for the manufacture of epoxy resins, ion-exchange
resins and other products), dimethyl formamide (used,
for example, as a solvent, reaction medium and interme-
diate in the manufacture of pharmaceuticals), n-butyric
acid (used in beverages, ?avorings, fragrances and other
products) and barium titanate powder (used for the man-
ufacture of electronic capacitors).
3. Specialty or effect or functional chemicals:These are
purchased because of their effect (or function), rather
than their chemical composition. Examples are pharma-
ceuticals, pesticides, dyestuffs, perfumes and ?avorings.
Because commodity and ?ne chemicals tend to be pur-
chased on the basis of their chemical composition alone,
they are undifferentiated. For example, there is nothing to
choose between 99.9% benzene made by one manufacturer
and that made by another manufacturer, other than price
and delivery issues. On the other hand, specialty chemicals
tend to be purchased on the basis of their effect or function
and are therefore differentiated. For example, competitive
pharmaceutical products are differentiated according to the
ef?cacy of the product, rather than chemical composition.
An adhesive is purchased on the basis of its ability to stick
things together, rather than its chemical composition and
so on.
Chemical Process Design and Integration R. Smith
? 2005 John Wiley & Sons, Ltd ISBNs: 0-471-48680-9 (HB); 0-471-48681-7 (PB)
However, undifferentiated and differentiated should be
thought of as relative terms rather than absolute terms for
chemical products. In practice, chemicals do not tend to
be completely undifferentiated or completely differentiated.
Commodityand?nechemicalproductsmighthaveimpurity
speci?cations as well as purity speci?cations. Traces of
impurities can, in some cases, give some differentiation
between different manufacturers of commodity and ?ne
chemicals. For example, 99.9% acrylic acid might be
considered to be an undifferentiated product. However,
traces of impurities, at concentrations of a few parts per
million, can interfere with some of the reactions in which
it is used and can have important implications for some
of its uses. Such impurities might differ between different
manufacturing processes. Not all specialty products are
differentiated. For example, pharmaceutical products like
aspirin (acetylsalicylic acid) are undifferentiated. Different
manufacturers can produce aspirin and there is nothing to
choose between these products, other than the price and
differentiation created through marketing of the product.
Scale of production also differs between the three classes
of chemical products. Fine and specialty chemicals tend
to be produced in volumes less than 1000 t·y
-1
.Onthe
other hand, commodity chemicals tend to be produced in
much larger volumes than this. However, the distinction
is again not so clear. Polymers are differentiated products
because they are purchased on the basis of their mechanical
properties, but can be produced in quantities signi?cantly
higher than 1000 t·y
-1
.
When a new chemical product is ?rst developed, it
can often be protected by a patent in the early years of
commercial exploitation. For a product to be eligible to
be patented, it must be novel, useful and unobvious. If
patent protection can be obtained, this effectively gives
the producer a monopoly for commercial exploitation of
the product until the patent expires. Patent protection lasts
for 20 years from the ?ling date of the patent. Once the
patent expires, competitors can join in and manufacture the
product. If competitors cannot wait until the patent expires,
then alternative competing products must be developed.
Another way to protect a competitive edge for a new
product is to protect it by secrecy. The formula for Coca-
Cola has been kept a secret for over 100 years. Potentially,
there is no time limit on such protection. However, for
the protection through secrecy to be viable, competitors
must not be able to reproduce the product from chemical
analysis. This is likely to be the caseonly for certain classes
of specialty and food products for which the properties of
2 The Nature of Chemical Process Design and Integration
the product depend on both the chemical composition and
the method of manufacture.
Figure 1.1 illustrates different product life cycles
1,2
.The
general trend is that when a new product is introduced
into the market, the sales grow slowly until the market
is established and then more rapidly once the market is
established. If there is patent protection, then competitors
will not be able to exploit the same product commercially
until the patent expires, when competitors can produce the
same product and take market share. It is expected that
competitive products will cause sales to diminish later in
the product life cycle until sales become so low that a
company would be expected to withdraw from the market.
In Figure 1.1, Product A appears to be a poor product that
has a short life with low sales volume. It might be that it
cannot compete well with other competitive products, and
alternative products quickly force the company out of that
business. However, a low sales volume is not the main
criterion to withdraw from the market. It might be that
a product with low volume ?nds a market niche and can
be sold for a high value. On the other hand, if it were
competing with other products with similar functions in
the same market sector, which keeps both the sale price
and volume low, then it would seem wise to withdraw
from the market. Product B in Figure 1.1 appears to be
a better product, showing a longer life cycle and higher
sales volume. This has patent protection but sales decrease
rapidly after patent protection is lost, leading to loss of
market through competition. Product C in Figure 1.1 is
a still better product. This shows high sales volume with
the life of the product extended through reformulation of
the product
1
. Finally, Product D in Figure 1.1 shows a
product life cycle that is typical of commodity chemicals.
Commodity chemicals tend not to exhibit the same kind
of life cycles as ?ne and specialty chemicals. In the early
years of the commercial exploitation, the sales volume
grows rapidly to a high volume, but then does not decline
and enters a mature period of slow growth, or, in some
exceptional cases, slow decline. This is because commodity
chemicals tend to havea diverse rangeof uses. Even though
competition might take away some end uses, new end uses
are introduced, leading to an extended life cycle.
The different classes of chemical products will have
very different added value (the difference between the
selling price of the product and the purchase cost of
raw materials). Commodity chemicals tend to have low
added value, whereas ?ne and specialty chemicals tend to
have high added value. Commodity chemicals tend to be
produced in large volumes with low added value, while
?ne and specialty chemicals tend to be produced in small
volumes with high added value.
Because of this, when designing a process for a
commodity chemical, it is usually important to keep
operating costs as low as possible. The capital cost of the
process will tend to be high relative to a process for ?ne or
specialty chemicals because of the scale of production.
When designing a process for specialty chemicals,
priority tends to be given to the product, rather than to
the process. This is because the unique function of the
product must be protected. The process is likely to be
small scale and operating costs tend to be less important
than with commodity chemical processes. The capital
cost of the process will be low relative to commodity
chemical processes because of the scale. The time to
Product
Sales
(t • y
-1
)
Product D 
Product C 
Product B 
Product A 
Patent Expiry
Product
Reformulation
Time
(y )
Figure 1.1 Product life cycles. (Adapted from Sharratt PN, 1997, Handbook of Batch Process Design, Blackie Academic and
Professional by permission).
Formulation of the Design Problem 3
market the product is also likely to be important with
specialty chemicals, especially if there is patent protection.
If this is the case, then anything that shortens the time
from basic research, through product testing, pilot plant
studies, process design, construction of the plant to product
manufacture will have an important in?uence on the overall
project pro?tability.
All this means that the priorities in process design are
likely to differ signi?cantly, depending on whether a pro-
cess is being designed for the manufacture of a commodity,
?ne or specialty chemical. In commodity chemicals, thereis
likely to berelatively little productinnovation, butintensive
process innovation. Also, equipment will be designed for a
speci?c process step. On the other hand, the manufacture
of ?ne and specialty chemicals might involve:
• selling into a market with low volume,
• short product life cycle,
• a demand for a short time to market, and therefore, less
time is available for process development, with product
and process development proceeding simultaneously.
Because of this, the manufacture of ?ne and specialty
chemicals is often carried out in multipurpose equipment,
perhaps with different chemicals being manufactured in
the same equipment at different times during the year.
The life of the equipment might greatly exceed the life
of the product.
The development of pharmaceutical products is such that
high-quality products must be manufactured during the
development of the process to allow safety and clinical
studies to be carried out before full-scale production.
Pharmaceutical production represents an extreme case
of process design in which the regulatory framework
controlling production makes it dif?cult to make process
changes, even during the development stage. Even if
signi?cant improvements to processes for pharmaceuticals
can be suggested, it might not be feasible to implement
them, as such changes might prevent or delay the process
from being licensed for production.
1.2 FORMULATION OF THE DESIGN
PROBLEM
Before a process design can be started, the design prob-
lem must be formulated. Formulation of the design problem
requires a product speci?cation. If a well-de?ned chemical
product is to be manufactured, then the speci?cation of the
productmightappearstraightforward(e.g.apurifyspeci?ca-
tion). However,if aspecialty productis to bemanufactured,
it is the functional properties that are important, rather than
the chemical properties, and this might require a product
design stageinordertospecifytheproduct
3
.Theinitialstate-
mentofthedesignproblemisoftenillde?ned.Forexample,
the design team could be asked to expand the production
capacityof anexisting plantthatproducesa chemicalthatis
a precursor to a polymer product, which is also produced by
the company. This results from an increase in the demand
for the polymer product and the plant producing the precur-
sor currently being operated at its maximum capacity. The
designer might well be given a speci?cation for the expan-
sion. For example, the marketing department might assess
that the market could be expanded by 30% over a two-year
period, which would justify a 30% expansion in the process
for the precursor. However, the 30% projection can easily
be wrong. The economic environment can change, leading
to the projected increase being either too large or too small.
It might also be possible to sell the polymer precursor in the
market to other manufacturers of the polymer and justify an
expansionevenlargerthan30%.Ifthepolymerprecursorcan
besoldinthemarketplace,isthecurrentpurityspeci?cation
of the company suitable for the marketplace? Perhaps the
marketplace demands a higher purity than what is currently
the company speci?cation. Perhaps the current speci?cation
is acceptable,but if the speci?cation could be improved, the
product could be sold for a higher value and/or at a greater
volume.Anoption might betonotexpandtheproductionof
thepolymerprecursorto30%,butinsteadtopurchaseitfrom
the market. If it is purchased from the market, is it likely to
be up to the company speci?cations, or will it need some
puri?cation before it is suitable for the company’s polymer
process? How reliable will the market source be? All these
uncertainties are related more to market supply and demand
issues than to speci?c process design issues.
Closer examination of the current process design might
lead to the conclusion that the capacity can be expanded
by 10% with a very modest capital investment. A
further increase to 20% would require a signi?cant capital
investment, but an expansion to 30% would require an
extremely large capital investment. This opens up further
options. Should the plant be expanded by 10% and a
market source identi?ed for the balance? Should the plant
be expanded to 20% similarly? If a real expansion in the
market place is anticipated and expansion to 30% would
be very expensive, why not be more aggressive and instead
of expanding the existing process, build an entirely new
process? If a new process is to be built, then what should
be the process technology? New process technology might
have been developed since the original plant was built
that enables the same product to be manufactured at a
much lower cost. If a new process is to be built, where
should it be built? It might make more sense to build
it in another country that would allow lower operating
costs, and the product could be shipped back to be fed
to the existing polymer process. At the same time, this
might stimulate the development of new markets in other
countries, in which case, what should be the capacity of the
new plant?
Fromallof these options, thedesign teammust formulate
a number of plausible design options. Thus, from the initial
4 The Nature of Chemical Process Design and Integration
ill-de?ned problem, the design team must create a series of
very speci?c options and these should then be compared
on the basis of a common set of assumptions regarding, for
example, raw materials prices and product prices. Having
speci?ed an option, this gives the design team a well-
de?ned problem to which the methods of engineering and
economic analysis can be applied.
In examining a design option, the design team should
start out by examining the problem at the highest level,
in terms of its feasibility with the minimum of detail to
ensure the design option is worth progressing
4
.Istherea
large difference between the value of the product and the
cost of the raw materials? If the overall feasibility looks
attractive, then more detail can be added, the option re-
evaluated,furtherdetailadded,andsoon.Byproductsmight
play a particularly important role in the economics. It might
be that the current process produces some byproducts that
can be sold in small quantities to the market. But, as the
process is expanded, there might be market constraints for
the new scale of production. If the byproducts cannot be
sold, how does this affect the economics?
If the design option appears to be technically and eco-
nomically feasible, then additional detail can be considered.
Material and energy balances can be formulated to give
a better de?nition to the inner workings of the process
and a more detailed process design can be developed. The
design calculations for this will normally be solved to a
high level of precision. However, a high level of preci-
sion cannot usually be justi?ed in terms of the operation of
the plant after it has been built. The plant will almost never
workpreciselyatitsoriginaldesign?owrates,temperatures,
pressures and compositions. This might be because the raw
materials are slightly different than what is assumed in the
design. The physical properties assumed in the calculations
might have been erroneous in some way, or operation at the
original design conditions might create corrosion or foul-
ing problems, or perhaps the plant cannot be controlled
adequately at the original conditions, and so on, for a mul-
titude of other possible reasons. The instrumentation on
the plant will not be able to measure the ?owrates, tem-
peratures, pressures and compositions as accurately as the
calculations performed. High precision might be required
for certain speci?c parts of the design. For example, the
polymer precursor might need certain impurities to be very
tightly controlled, perhaps down to the level of parts per
million.Itmightbethatsomecontaminantinawastestream
might be exceptionally environmentally harmful and must
be extremely well de?ned in the design calculations.
Even though a high level of precision cannot be justi?ed
in many cases in terms of the plant operation, the design
calculations will normally be carried out to a reasonably
high level of precision. The value of precision in design
calculationsisthattheconsistencyofthecalculationscanbe
checkedtoallowerrorsorpoorassumptionstobeidenti?ed.
It also allows the design options to be compared on a valid
like-for-like basis.
Because of all the uncertainties in carrying out a design,
thespeci?cationsareoftenincreasedbeyondthoseindicated
by the design calculations and the plant is overdesigned,
or contingency is added, through the application of safety
factors to the design. For example, the designer might
calculate the number of distillation plates required for a
distillation separation using elaborate calculations to a high
degree of precision, only to add an arbitrary extra 10% to
the number of plates for contingency. This allows for the
feed to the unit not being exactly as speci?ed, errors in the
physical properties, upset conditions in the plant, control
requirements, and so on. If too little contingency is added,
the plant might not work. If too much contingency is added,
the plant will not only be unnecessarily expensive, but too
much overdesign might make the plant dif?cult to operate
and might lead to a less ef?cient plant. For example, the
designer might calculate the size of a heat exchanger and
then add in a large contingency and signi?cantly oversize
the heat exchanger. The lower ?uid velocities encountered
bytheoversizedheatexchangercancauseittohaveapoorer
performance and to foul up more readily than a smaller
heat exchanger. Thus, a balance must be made between
different risks.
Insummary,theoriginalproblemposedtoprocessdesign
teams is often ill-de?ned, even though it might appear to
be well de?ned in the original design speci?cation. The
designteammustthenformulateaseriesofplausibledesign
options to be screened by the methods of engineering and
economic analysis. These design options are formulated
into very speci?c design problems. Some design options
mightbeeliminatedearlybyhigh-levelargumentsorsimple
calculations. Others will require more detailed examination.
In this way, the design team turns the ill-de?ned problem
into a well-de?ned one for analysis. To allow for the many
unquanti?able uncertainties, overdesign is used. Too little
overdesign might lead to the plant not working. Too much
overdesign will lead to the plant becoming unnecessarily
expensive,andperhapsdif?culttooperateandlessef?cient.
A balance must be made between different risks.
Consider the basic features of the design of chemical
processes now.
1.3 CHEMICAL PROCESS DESIGN
AND INTEGRATION
In a chemical process, the transformation of raw materials
into desired chemical products usually cannot be achieved
in a single step. Instead, the overall transformation is bro-
ken down into a number of steps that provide intermediate
transformations.Thesearecarriedoutthroughreaction,sep-
aration, mixing, heating, cooling, pressure change, particle
size reduction or enlargement. Once individual steps have
been selected, they must be interconnected to carry out the
The Hierarchy of Chemical Process Design and Integration 5
Feed
Streams
Feed
Streams
Product
Streams
Product
Streams
(a)  Process design starts with the synthesis of a process to convert raw
 materials into desired products.
(b)  Simulation predicts how a process would behave if it was
       constructed.
d.
?
Figure 1.2 Synthesis is the creation of a process to transform
feed streams into product streams. Simulation predicts how it
would behave if it was constructed.
overall transformation (Figure 1.2a). Thus, the synthesis of
a chemicalprocess involves two broad activities. First, indi-
vidual transformation steps are selected. Second, these indi-
vidualtransformationsareinterconnectedtoformacomplete
process that achieves the required overall transformation. A
?owsheet is a diagrammatic representation of the process
steps with their interconnections.
Once the ?owsheet structure has been de?ned, a
simulation of the process can be carried out. A simulation
is a mathematical model of the process that attempts to
predict how the process would behave if it were constructed
(Figure 1.2b). Having created a model of the process, the
?owrates, compositions, temperatures and pressures of the
feeds can be assumed. The simulation model then predicts
the ?owrates, compositions, temperatures, and pressures
of the products. It also allows the individual items of
equipment in the process to be sized and predicts, for
example, how much raw material is being used or how
much energy is being consumed. The performance of the
design can then be evaluated. There are many facets to the
evaluation of performance. Good economic performance is
anobvious ?rstcriterion,but itis certainlynotthe only one.
Chemical processes should be designed as part of a
sustainable industrial activity that retains the capacity of
ecosystems to support both life and industrial activity into
the future. Sustainable industrial activity must meet the
needs of the present, without compromising the needs of
future generations. For chemical process design, this means
that processes should use raw materials as ef?ciently as is
economic and practicable, both to prevent the production of
waste that can be environmentally harmful and to preserve
the reservesof raw materialsas much aspossible. Processes
should use as little energy as is economic and practicable,
both to prevent the build-up of carbon dioxide in the
atmosphere from burning fossil fuels and to preserve the
reserves of fossil fuels. Water must also be consumed in
sustainable quantities that do not cause deterioration in the
quality of thewatersourceandthe long-termquantity ofthe
reserves. Aqueous and atmospheric emissions must not be
environmentally harmful, and solid waste to land?ll must
be avoided.
The process must also meet required health and safety
criteria. Start-up, emergency shutdown and ease of control
are other important factors. Flexibility, that is, the ability
to operate under different conditions, such as differences
in feedstock and product speci?cation, may be important.
Availability, that is, the number of operating hours per
year, may also be critically important. Uncertainty in the
design, for example, resulting from poor design data, or
uncertainty in the economic data, might guide the design
away from certain options. Some of these factors, such as
economic performance, can be readily quanti?ed; others,
such as safety, often cannot. Evaluation of the factors that
are not readily quanti?able, the intangibles, requires the
judgment of the design team.
Once the basic performance of the design has been eval-
uated, changes can be made to improve the performance;
the process is optimized. These changes might involve the
synthesis of alternative structures, that is, structural opti-
mization. Thus, the process is simulated and evaluated
again, and so on, optimizing the structure. Alternatively,
each structure can be subjected to parameter optimization
by changing operating conditions within that structure.
1.4 THE HIERARCHY OF CHEMICAL
PROCESS DESIGN AND
INTEGRATION
Consider the process illustrated in Figure 1.3
5
. The process
requires a reactor to transform the FEED into PRODUCT
(Figure 1.3a). Unfortunately, not all the FEED reacts.Also,
part of the FEED reacts to form BYPRODUCT instead of
the desired PRODUCT. A separation system is needed to
isolate the PRODUCT at the required purity. Figure 1.3b
shows one possible separation system consisting of two
distillation columns. The unreacted FEED in Figure 1.3b
is recycled, and the PRODUCT and BYPRODUCT are
removed from the process. Figure 1.3b shows a ?owsheet
where all heating and cooling is provided by external
utilities (steam and cooling water in this case). This
?owsheetisprobablytoo inef?cientinits useofenergy,and
heat would be recovered. Thus, heat integration is carried
out to exchange heat between those streams that need to be
cooled and those that need to be heated. Figure 1.4
5
shows
two possible designs for the heat exchanger network, but
many other heat integration arrangements are possible.
The ?owsheets shown in Figure 1.4 feature the same
reactor design. It could be useful to explore the changes in
reactor design. For example, the size of the reactor could
be increased to increase the amount of FEED that reacts
5
.
Read More
Offer running on EduRev: Apply code STAYHOME200 to get INR 200 off on our premium plan EduRev Infinity!

Content Category

Related Searches

practice quizzes

,

Important questions

,

shortcuts and tricks

,

Extra Questions

,

Exam

,

MCQs

,

Summary

,

past year papers

,

mock tests for examination

,

Free

,

study material

,

The Nature of Chemical Process Design and Integration - Chemical Engineering Chemical Engineering Notes | EduRev

,

The Nature of Chemical Process Design and Integration - Chemical Engineering Chemical Engineering Notes | EduRev

,

The Nature of Chemical Process Design and Integration - Chemical Engineering Chemical Engineering Notes | EduRev

,

pdf

,

Viva Questions

,

ppt

,

video lectures

,

Objective type Questions

,

Previous Year Questions with Solutions

,

Sample Paper

,

Semester Notes

;