Chemical Engineering August 2016 - 42

106
10-1
100
CAPEX per
1 kg/h
105
104
103
Loop reactor
102
Stirred-tank reactor
101
102
103
104
105
Cost per unit of reactor volume ( 1,000 €/m3)
FIGURE 3. Shown here are
benchmarking metrics for
industrial microreactors
flow of electric current can produce visible
light was found by Davy in 1802, but it took
another 75 years before Edison made it a
commercial success. Better vacuum-pump
technology and filament material needed to
be developed first.
Other useful insights can be found in the
FIGURE 4. This schematic
shows the secondary flow
phenomena that occurs perpendicular
to the flow direction
in a curved channel
Primary
flow
diffusion-of-innovation theory of Rogers [8],
where the adoption of a new technology
is explained in a more behavioral context.
The rate of adoption into a community can
be described by the difference in behavior
between five categories of adopters. The
degree of adoption over time (since the invention
was made) follows a bell-curve. In
the beginning, the adoption is driven by the
technology enthusiasts and visionaries who
are willing to take a risk, and motivated by
newness rather than robustness. Adoption
rate is low and a lot of the prototyping
fails with the subsequent shake-out of the
start-up companies. The expression of the
technology can be hampered by missing
complementary factors. The inflated expectations
are brought back to reality, and without
further development, the actual adoption
can even be stalled.
A-A
A-A
Dean vortices (secondary flow)
A-A
A-A
Steady
Chaotic in
time and
position
Increasing
* curvature
* velocity
42
Hurdles toward industrial application
Let's go back to the adoption of microreactor
technology in the chemical industry. The
early success in the laboratory demonstrated
the potential of the technology and created
high expectations. The product of laboratory-scale
experiments, however, is information,
and for this we don't need volume. A
microliter-sized reactor can tell us about the
reaction kinetics and the achievable selectivity,
or produce a few grams of substance for
a library. The microreactor may be a costly
prototype as long as the value of the information
is satisfactory. Things change when
the product is mass, when tons of substance
are required, when operation has to
meet plant and business requirements.
The first hurdle we perceived was that
of scaleup to a sufficiently large volume to
deliver the required production capacity. Although
productivity in a microreactor can
reach up to 100,000 kg/m3h, the required
reactor volume is still substantial and a single
channel will not be sufficient. Astronomically
high numbers of parallel channels would be
required (Figure 2). Numbering-up is often
hailed as the solution with the added benefit
of avoiding scaleup effects. In theory this
is true, but the practical implications on the
technical feasibility (manifolding) are enormous,
let alone the impact on the economics
of the reactor construction. This problem
poses a true dilemma. The performance of
a microreactor comes from its micrometersized
channel dimension, and traditional
scaling-up would inevitably result in loss of
that performance, bringing us back to conventional
tube reactors.
The second hurdle that we encountered
was the manufacturability and its impact on
the cost level. In the early 2000s, with the
technology still in its infancy, it was difficult to
find a technology provider who could solve
the numbering-up challenge on a technical
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
AUGUST 2016
106
Microreactor
101
102
capacity
(1,000 €)
Moore [9] expanded the model by focusing
on this critical point (chasm) in the
bell-curve between early adopters and early
majority (Figure 1). The most difficult step is
making this transition because the expectations
of the adopters change. The early
majority is more pragmatic in their decision.
They need to be convinced by real benefits
and performance. A lack of technology
readiness is not accepted. Both economic
and technical feasibility need to reach a satisfying
level. In our opinion this is the true
reason for the reluctance to implement industrial
microreactors.
Productivity (kg/m3h)
http://WWW.CHEMENGONLINE.COM
Chemical Engineering August 2016

Table of Contents for the Digital Edition of Chemical Engineering August 2016
