education for chemical engineers 1 4 ( 2 0 1 6 ) 43–48

Contents lists available at ScienceDirect

Education for Chemical Engineers

journal homepage: www.elsevier.com/locate/ece

Teaching chemical product design

Alirio Rodrigues a , E.L. Cussler b,∗

a University of Porto, Praça de Gomes Teixeira, 4099-002 Porto, Portugal
b CEMS, University of Minnesota, Minneapolis, MN 55455, USA

a r t i c l e i n f o a b s t r a c t

Article history: The chemical industry today includes both commodity products and higher value-added
Received 6 August 2015 products. While the design of commodities is dominated by the process costs, higher value-
Received in revised form 23 added products also depend on product design, including discovery, product selection,
December 2015 and time-to-market. Chemical engineering education has sensibly begun to change toward
Accepted 24 December 2015 courses on both process and product design. However, while there is an emerging consen-
Available online 31 December 2015 sus that these changes should take place, there is no clear agreement on what the changes
should be. Moreover, these new directions are very difficult to teach, at least in the current
environment. This paper will discuss different efforts to incorporate product design into the
chemical engineering curriculum and different successes in doing so. However, while the
value of including this material seems unquestioned, the way in which it is best taught is
unclear.
© 2016 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. The current chemical industry
Unlike the hopes expressed by some industrial leaders, the
chemical business is not a growth business. It was a growth
business 50 years ago, when the manufacturers of tex-
tile fibers changed the clothes that were on the backs of
all the peoples in the world. At that time, the chemical
business had growth rates that were comparable with the
electronic business of today. That clearly is no longer the
case. Over at least last 25 years, the chemical industry has
been a commodity business, using hydrocarbon feedstocks
obtained from less developed, frequently unstable areas. As
a result, chemical corporate planning has become extremely
difficult.
One result was that the chemical business went into a
period of contraction. Larger chemical companies began sell-
ing off their commodity products. The Dupont Company sold
nylon, its signature product, to Koch for $4.4 billion. Gen-
eral Electric sold its plastics business to Sabic. Dow Chemical
tried to sell most of its commodity chemical business to the
Kuwaitis. One month after they had signed the contract with
Dow, the Kuwaitis decided they had made a bad deal, and
paid over a billion dollars to extricate themselves from their
contract. These are not signs of a growth business.
And then came fracking.
Fracking is a technique of horizontal drilling underground.
Developed over 50 years ago as a means of enhancing produc-
tion of conventional wells, fracking in its modern incarnation
involves not only horizontal drilling, but subterranean explo-
sions to open reservoir formations. When these formations
are opened, they then are subjected to high pressure injec-
tions of aqueous suspensions which include sand. The grains
of sand prop cracks in the formation open, allowing petroleum
liquids and natural gas to be harvested. This increased produc-
tion of natural gas has caused natural gas prices to drop almost
in half, giving the chemical commodity business in developed
countries has a new lease on life.
Now, the chemical business is undergoing a major renais-
sance. The resulting changes are magnified because the
business had anticipated a static market, at least within the
developed world. The business had been cutting back on hires,
so that the average age of the employees was much greater
than it had been historically. In some cases, more than half of
the company employees were retirement eligible. The cheaper

∗
Corresponding author. Tel.: +1 612 625 1596.
E-mail addresses: arodrig@fe.up.pt (A. Rodrigues), cussler@umn.edu (E.L. Cussler).
http://dx.doi.org/10.1016/j.ece.2015.12.001
1749-7728/© 2016 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
44 education for chemical engineers 1 4 ( 2 0 1 6 ) 43–48

natural gas suddenly changed the industries’ goals. If they

Table 1 – What will future products be? To think about
now plan to make commodities, they will need to hire younger this, make a list of potential products, and decide how
people. Thus there will be new jobs for engineers to work on their amounts and values change which aspects are
commodities. important.
But at the same time, many chemical companies believe Polypropylene Nylon
that in the long term they have a more significant opportunity Petrol Vanillin from black liquor
in products other than commodities (Cussler and Moggridge, Smart labels Light emitting diodes
2011; Wesselingh et al., 2007; Wei, 2007; Teixeira et al., 2013; Biofuels from biomass CO2 capture
Moggridge et al., 2006). The reason is simple: non-commodity Sulfuric acid Solar shingles
Thermopane windows Freon-free foam
products can have significantly higher added value. Corpora-
Oxygen Perfume engineering
tions planning new products will also need new people, so
TiO2 Epoxy resins
there will be a second market for engineers to work on non- Silky textiles Penicillin
commodity products. While the relative size of commodity
and non-commodity employment markets is unknown, many
expect the numbers in the two areas to be roughly compara- Table 2 – How to think about future products to explore
ble. This altered product mix was recognized in 2005 by the this, we suggest that you define a fourth category, and
identify its characteristics.
European Federation of Chemical Engineering (EFCE) with the
statement: Commodities Molecules Microstructures

Key Cost Speed Function

“While all chemical engineers still need much of the
traditional skills, the European Federation of Chemical
Engineering feels there is now a need to include some
knowledge of “product engineering” in the common core”.
Thus we faculty need to think of educating our students for
both these business areas. How we do so for non-commodity
products is the subject of this article.
2. Thinking about chemical products
Some think that the design of chemical products should not
be that different than the design of chemical processes. This
is not true (Ulrich and Eppinger, 2008). As an illustration, con-
sider the products shown in Table 1, products that we have
been asked about in the last few years. Polypropylene, sulfu-
ric acid, and oxygen are chemical products. In each case, the
processes by which these are made are the key to their devel-
opment. But there are many other chemical products in Table 1
that will require different development. Three examples are
shown in boldface in the table. With smart labels, we could
label a package of chicken to track the age and the temper-
ature history of the chicken. On our homes, we could install
solar shingles which not only protect the house but generate
electricity on the sunny days. We could seek textiles which feel
silky but which are made from completely different materi-
als than from silk. Each of these topics—smart labels, solar
shingles, and silky textiles—is a chemical product, but the
Basis Unit operations Chemistry Microstructure
Risk Feedstock Discovery Science
way in which they are developed is different than the ways by
which processes are developed processes for spinning nylon
or synthesizing titanium dioxide.
Once we recognize this group of products, we need a way
of thinking systematically about these. One way, described in
Teixeira et al. (2013), is to group them under three categories,
as listed in Table 2. Other product classifications can also be
valuable, but we can do better if we organize our thinking
around categories. If you have already taught product design,
we invite you to skip to the next section. If you have not taught
product design, you may wish to read the following summary
about product categories.
The three categories of products are commodities,
molecules, and microstructures. The way in which these
three categories of chemical products are developed is dif-
ferent. Commodities are the most familiar, because their
development depends so strongly on cost. Profit margins for
commodities are small, so low cost processing is an axiom of
classical chemical engineering ideas. Small profit margins are
the why the commodity chemical business has been threat-
ened by having feedstocks from less stable, less developed
countries.
The familiar process engineering, which is essential to
commodity chemicals, is much less important for molecular

Fig. 1 – Three molecular products. The discovery of these products is more important than the process by which they are
made.
 education for chemical engineers 1 4 ( 2 0 1 6 ) 43–48
Table 3 – Product design decides what to make.
Process design Product design
Batch vs. continuous Customer need
Input/output Idea generation
Recycles Selection
Separation/heat Manufacture
and microstructured products. To illustrate this, consider the
three molecular products shown in Fig. 1. Penicillin is an
antibiotic made by fermentation of an agricultural waste
stream, corn steep liquor. Prozac, one of the earliest effective
treatments for mental disorders, is made by chemical synthe-
sis. Premarin, which alleviates symptoms of menopause, is an
extract of the urine of pregnant mares. The molecular struc-
ture of any these is dramatically more complicated than chem-
ical products like oxygen, sulfuric acid, and polypropylene.
The important factors given in Table 2 for molecular prod-
ucts are completely different than those for commodities. The
key for molecular products is the time-to-market, that is, the
speed of development. Unit operations are less important than
chemical synthesis. The real risk is in discovery. While drug
companies often assert their products are expensive because
they must pay for research, they do not point out the research
of successful products is not unusually expensive. The prob-
lem is that the expensive research is done on drugs which turn
out to be unsuccessful.
The third category—microstructures—involves those prod-
ucts that gain their value by function that they provide. Paints,
cosmetics, flavorings, and tooth whitening strips are exam-
ples. This area is missing the fundamental science required
for product development. One successful example supplying
this missing science is the case of perfumes (Teixeira et al.,
2013), which underscores the opportunity in other areas like
food flavor and clothing comfort. This lack of integrated sci-
ence is why the EPSRC has invested in the development of
particulate processes at Leeds University.
3. Process versus product design
Now that we appreciate the differences between products, we
can begin to explore how differently the design of these differ-
ent types will proceed. One possible template for such designs
is shown in Table 3. The left hand side of that table gives a
sequence for process design. While this particular sequence is
that suggested by Douglas (1988), other design books suggest
similar sequences.
Douglas’s template has four steps. First, we decide whether
the process will batch or continuous. For commodity prod-
ucts, the answer almost always is continuous, to reduce the
capital costs per mass for products that have small profit mar-
gins. Second, we draw an input-output diagram, also called a
block diagram, which shows the flows in the process as arrows
between pieces of equipment, shown as simple boxes. The
third step is to consider recycles, including the separation and
reuse of unreacted reagents. Finally, we integrate separation
processes and minimize energy use. Ironically, this final step
tends to take the most time and detailed calculation, using
the effective software developed for the commodity chemical
industry.
The product design template shown in the right hand side
of Table 3 is completely different (Cussler and Moggridge, 2011;
Wesselingh et al., 2007). We begin with the customers’ needs.
45
Fig. 2 – The current skill set is adequate. Interestingly,
national differences in training do not dramatically affect
student performance in product design.
We then invent ideas to meet this need. Third, we select the
best few ideas for further development. Finally, we decide how
to manufacture the product, a step which includes all of the
sequence listed on the left hand side of Table 3. Thus process
design asks how we will make the product, but product design
must also decide what product we will make.
We now look at these four product design steps in more
detail to illustrate the thinking involved. As an example of
needs, we consider a new type of battery. A battery a small
package which supplies energy for some device, like a cell
phone, a laptop, or a soldier in the battlefield. Identifying this
need frequently involves marketing, an area which many engi-
neers disparage. However, those experienced in marketing will
rarely have enough technical knowledge to identify specific
attributes which a product must have. They can rarely write
quantitative specifications. This is why engineers should be
involved in the definition of product needs.
The second step in product design is the generation of
ideas. In order to have some good ideas, we typically need
about 100 ideas, most of which will be bad. In generating these
ideas, you must not in any way be critical, and explore how
these ideas can be generalized. We need a lot of ideas in order
to find a few good ones.
Third, we must select the best ideas for the development.
Because we will probably not be able to develop more than two
or three quantitatively, we must learn to quickly screen the
ideas, reducing the number down to perhaps 10. This type of
screening involves wisdom and the ability to make quick, qual-
itative judgments. It is an area where our students frequently
need practice.
When they start selecting among the last few ideas, they
often will not know how to proceed. Some trivialize this situa-
tion by saying that “We cannot compare apples and oranges.”
This is nonsense. Every time we go to the market to buy fruit,
we compare apples and oranges. Such comparisons are ones
that engineering students need to practice making.
The fourth step, the manufacture, is not reviewed here
because this already is discussed in process design courses.
4. Key steps in product design
At this point, we may wonder if our students actually have
the ability to carry out product design. We believe they do.
To support this belief, we plot chemical engineering skills on
a triangular diagram, as shown in Fig. 2. This graph ideal-
izes chemical engineering as a mixture of the three topics:
46 education for chemical engineers 1 4 ( 2 0 1 6 ) 43–48

Fig. 3 – For commodities, “Manufacture” is Key. The complexity results from low profit margins.

empirical principles, shown at the top of the diagram; con-

tinuum ideas, especially those derived from physics, showing
in the lower left hand corner; and chemical concepts, based
on molecules and shown in the lower right hand corner. Any
chemical engineering training is a mixture of these ideas. That
in the United Kingdom is biased toward empirical principles,
but that in Germany emphasizes physical ideas. The United
States is biased toward chemistry. Still, all three mixtures of
ideas produce students who are effective in the chemical busi-
ness.
There is no reason to believe that this mixture of skills
needs to be dramatically changed to extend chemical process
design to chemical product design. To be sure, chemical engi-
neers trained in the United Kingdom are going to be superior
for products which are pastes or suspensions because those Fig. 4 – For molecules, “Ideas” is Key. The drug Zolodex has
subjects are studied more completely in the UK. In the same a market of close to a billion dollars per year based on less
sense, American engineers may do better on projects that than 50 kg of product.
involve heavy doses of synthetic chemistry. But we do not
think these differences are significant in making judgments
about product design. oxygen by cryogenic distillation, or making propylene from
Irrespective of the background, we do believe the differ- a hydrocarbon feed stock are not going to have much appli-
ent steps in product design will have dramatically different cability for Zolodex. Here, “ideas,” not “manufacture”, is the
importance depending on the products being designed. For key.
commodity products, illustrated by the simplified process for For microstructured products, the situation is more compli-
making ammonia in Fig. 3, “manufacture” is clearly the key cated because these products vary more widely. One example
element. Indeed, this is why for the commodity based chemi- is the use of anti-icing fluids used on jet aircraft. These flu-
cal curriculum, often the only curriculum represented in edu- ids are polymer gels sprayed onto airplanes ready to take-off.
cation, commodity manufacturing is emphasized in the cap- These gels melt falling snow, preventing ice from forming on
stone design classes. Molecular products, on the other hand, the wings. The gels change the shape of the airfoil and hence
are dominated by “ideas”. An example, shown in Fig. 4, is the compromise its performance. However, because the gels are
drug Zolodex. This drug has net sales over $0.8 billion per year, Bingham fluids, the gel is stripped from the wing during take-
but those sales are based on less than 50 kg of material. Just off, so that a plane that may be only marginally able to fly at
imagine process control for a process which is producing only the start of the runway can fly by the time it reaches take-off
50 kg/year. Process control concepts developed for producing speed. In this case, making the “needs” specific is important.
 education for chemical engineers 1 4 ( 2 0 1 6 ) 43–48
Table 4a – Comparing product design courses. More detailed syllabi for courses in USA and UK are available from the
departments involved. Harvard’s MBA is included only for comparison.
Minnesota (2) Leeds
Taught to All 4th year All 3rd year
When Last term Last term
Total # lectures 35 22
Length class, weeks 14 11
# reports 1 4
Positives Promotes teamwork Promotes teamwork
Negatives Few students work hard Too much material
5. Product design is hard to teach
The rationale above demonstrates that product design should
be an important part of any chemical engineering curriculum.
However, though it has been taught for almost 20 years in a
variety of institutions, it has not spread broadly. We do not
think that its value is not appreciated or understood. This
is especially true in developed countries, which have more
to lose if the chemical business is restricted to commodities.
Nonetheless, product design has not become a core chemical
engineering subject. Why?
We believe that this is because product design is extremely
hard to teach. It does not involve the same degree of defini-
tion as thermodynamics or fluid flow. The material presented
in product design changes significantly every year. If student
reports are expected, the subject of those reports will usu-
ally change every year. Grading these reports takes enormous
time. This means that the amount of work required per faculty
member can be larger for product design than for every other
class.
Moreover, product design faculty must be comfortable
working on subjects where definite answers are not always
known (Lape et al., 2014). The solutions to any design project
may be different from professor to professor and from student
to student. These solutions may differ still more dramatically
from one university to the next, even when the product is
the same. But engineering faculty are often more comfortable
with definite quantitative answers, which do not often occur
in product design. We understand this. We are in this position,
too. But we still feel that the importance of the subject means
that we should do our best to include product design as part
of chemical engineering education.
The fact that the subject is hard to teach is reflected on
the content of different courses on this subject. These courses
fall into three general formats. The first format tries to bend
the process design course into a product orientation. It does
so by beginning with a product, like a coffeemaker or a small
pyrolysis reactor to make liquid bio-fuel for a single farm. In
doing so, it skips the needs and ideas steps of the template pre-
sented above. This way of teaching often pre-selects one or two
alternative technologies and then applies conventional pro-
cess design tools to optimizing the product that is produced.
The course is reasonably successful.
We feel this strategy dilutes the essentials of product
design and does not fulfill our more general objectives. Rather
than asking students to think in different areas, it encourages
extending process design techniques to a more product-
oriented area. In so doing, it is effective; but we believe that
it is not going help students dealing with that part of the
47
Cambridge Minnesota (1) Harvard business
All 4th year All 4th year Postgraduates
First term Last term Four terms
24 35 180
8 7 60
4 (oral) 6 30
Promotes teamwork All students try Promotes teamwork
hard
Final reports weak Does not Limited by special cases
promote
teamwork
chemical business that switches toward high value-added
products.
The other two formats for product design courses are illus-
trated by five examples, in US and UK universities in Table 4a,
and in Portuguese universities in Table 4b. Most of these
courses, shown in columns 2 and 3 of the tables, use 20–35
one hour lectures as a guide for students assigned a single
project. The students work in teams of three to five persons
to study and report, both orally and in writing. The students
may be guided by adjunct faculty who are experienced in the
product area being studied. Often, students are anxious about
how they will be graded. The permanent faculty rarely spend
more time than the students in developing a solution to the
projects, so that faculty evaluation tends to be high level, with-
out any detailed knowledge specific to the project. Every year,
one or two projects are enormously successful, generating
major interest in students, industrial sponsors, and venture
capitalists; but the majority of the projects are adequate but
unexciting. Anecdotally, the assessed value of the class is felt
by the students to be average when the students graduate; but
the assessed value is judged much greater, probably greater
than any other course, for students five years after gradua-
tion. Still, to our knowledge, there is no objective study of how
much is learned in these courses, the way that there is for a
flipped classroom (Saraiva and Costa, 2004).
The experiences in Portugal parallel those at other institu-
tions. Product design teaching tends to be led by one professor
with an evangelical commitment to the subject, followed by
others sensibly innovating with the format. For example, in
Coimbra, it was Pedro Saraiva (Moggridge et al., 2006; Saraiva
and Costa, 2004). At Porto, the lecturer gives only makes an
introduction and a few lectures on patents and economic eval-
uation. Instead, the students present in one class every week
a chapter of one assigned books; they review progress on their
projects in a second weekly class. Each student team ends
the course with a final presentation. This format insures that
students remain engaged and cannot procrastinate.
A second product design course format, shown in the next
to the last column of Table 4, sacrifices the gains of working in
teams for more uniform student effort and experience. While
the number of classes is the same, the number of projects
equals the number of weeks in the course. As a result, the
projects must be much more structured, specific to part of the
design template, and illustrating one specific aspect of product
design. The weekly projects are developed in detail by the fac-
ulty, so that the control over what is learned is more complete.
While we believe that this format produces better student
engagement, we recognize that it requires a still greater invest-
ment of faculty time, and so is rarely practiced, even by us.
48 education for chemical engineers 1 4 ( 2 0 1 6 ) 43–48
Table 4b – Comparing product design courses in Portugal. More detailed syllabi for courses are available from the
departments involved.
Coimbra
Taught to 4th year
When Second semester
Total # lectures 42 lectures 35 classes on practice
Length class, weeks 14 weeks
# Reports 1 plus 3 presentations
Positives Integration of knowledge
Interactive teamwork
Free choice of projects
Negatives Problems chosen can be too complex
Those who have not taught this type of course sometimes
worry that our engineering students are missing some “soft”
skills, sometimes supplied by business courses. In our experi-
ence, we have not found our students are missing anything
essential. One student group even developed a website for
managing the information of the Product Engineering course.
But these different course formats lead us to the ques-
tion of what the ideal product design course should be. In
other words, if we faculty were without other responsibili-
ties except teaching chemical product design, what course
would we teach? Given this freedom, we are drawn to the
case study method used for the Master of Business Admin-
istration (MBA) at the Harvard Business School, summarized
in the last column of Table 4. There, over 200 faculty wrote
250 case studies of specific business situations per year. These
case studies are the faculty’s research. The case studies then
serve as the core of instruction for the MBA. Students begin by
reading individual studies. They then work in teams to resolve
the problems which the studies pose. Their resolutions are
explored by faculty and other students in discussion with the
entire class. This method is judged by the business community
to be extremely successful.
This method could work in engineering if we invented a
mechanism for setting effective problems. While it is hard
to imagine how one university could do this every year, a
group of universities could possibly cooperate to assemble a
set of projects. One reviewer of this paper saw this as a major
challenge:
“Developing case studies is challenging and therefore
off-putting to potential instructors. The authors should
directly address how much of a detriment it would be to
student learning were cases to be re-used in subsequent
years. This is clearly institution dependent (for example,
honor codes at some institutions could be used to sti-
fle students passing on solutions). . .. Given the potentially
wide design space where a large number of feasible design
alternatives could be generated, it seems that it might be
possible to reuse some cases regardless of institutional
environment.”
We agree that reusing problems is a major way to sus-
tain sound teaching at a minor risk of compromised student
Porto
4th year
First semester
Few lectures work in four student teams
14 weeks
1 written, weekly oral
Promotes teamwork
Develops “soft skills”
Some weaknesses in final exam
evaluation. We also recognize that changing the details of
a specific project can encourage independent thought. Of
example, we have had projects on vanillin from lignin, on
skin care cream, on microcapsules for textiles, on perfume
design, and on lab on a chip. Any of these is easily gen-
eralized and extended. Examples in books on this subject
also provide parallel problems which give roadmaps toward
solutions. Moreover, we have one enormous advantage over
those working in business: what we teach fits into the general
intellectual framework provided by science and engineering.
We always emphasize that our results use tools that apply
broadly, and that our scientific laws and engineering corre-
lations can be used in situations beyond the specific ones
which we are studying. In contrast, business is obsessed with
the details which make the specific cases different. If we can
find the right format, we can offer better teaching of chemical
product design than the business schools can offer for case
studies.
References
Cussler, E.L., Moggridge, G.D., 2011. Chemical Product Design, 2nd
ed. Cambridge University Press, New York.
Douglas, J.M., 1988. The Conceptual Design of Chemical
Processes. McGraw-Hill Science/Engineering/Math, New York.
Lape, N.K., Levy, R., Yong, D., Haushalter, K., Eddy, R., Hankel, N.,
2014. Probing the inverted classroom: a controlled study of
teaching and learning outcomes in undergraduate
engineering and mathematics. In: Proceedings of the ASEE
Annual Conference, (paper ID #9475).
Moggridge, G., Costa, R., Saraiva, P., 2006. Chemical product
engineering: the future? TCE, 25–27 (November).
Saraiva, P., Costa, R., 2004. A chemical product design course with
quality focus. Chem. Eng. Res. Des. 82 (A11), 1474–1484.
Teixeira, M., Rodriguez, O., et al., 2013. Perfume Engineering:
Design, Performance and Classification. Elsevier
Butterworth-Heineman, Waltham, MA.
Ulrich, K.T., Eppinger, S.D., 2008. Product Design and
Development, 4th ed. McGraw-Hill Education, New York.
Wei, J., 2007. Product Engineering: Molecular Structure and
Properties. Oxford University Press, New York.
Wesselingh, J.A., Kiil, S., Vigild, M.E., 2007. Design & Development
of Biological, Chemical, Food, and Pharmaceutical Products.
Wiley, Hoboken, NJ.