Chemical Engineering June 2020 - 34

and are largely ignored in most asu design
calculations. the volumes in table 1
illustrate what is contained in the air feeding
an asu that produces 100 tons per
day of oxygen. this is a common size for
a plant that could easily make all of its
oxygen product as a liquid. if all product
was produced as liquid, it would be over
7.6 million gal/yr. at 100% recovery, the
ne production would be less than 400
gal/yr, while Kr would be around 50 gal/
yr and xe just under 5 gal/yr. rare gases
are typically sold as gaseous liters at
standard conditions. one liter equals
0.001 of a cubic meter, which tends to
make the volumes seem larger.
the potential amount of rare gases
that could be produced from a 100 ton/d
asu plant is too small to support the
capital needed for systems needed for
the isolation of rare gases. However, rare
gases can be valuable byproducts, so it
can be attractive to collect and refine one
or more of them at large asus.
product demand and projected pricing
determine at which point it becomes
economically
attractive
to
include
rare
gases as part of the product slate. a airseparation
plant that produces nominally
1,000 ton/d of oxygen is usually the point
rare gas recovery starts to become attractive.
new asus are being built as large as
5,500 ton/d (5,000 metric tons per day).
Historically, the prices of rare gases
have been quite elastic with supply and
demand. When demand increases,
the prices increase and more capacity
is added to the market, which causes
downward pressure on the price. this
situation was evident in 2015, when a
variety of factors disrupted the ne market,
causing prices to rise by a factor of
six or more. prices have since returned
to historical levels.
the decision to produce rare gases
often relates to the value of the downstream
product, since oxygen (or nitrogen,
or both) often represents only a
few percent of the raw material costs to
make the final product. producing rare
gases may not have enough impact on
the cost to of the final product to justify
any additional costs or risks associated
with including them. the risks include
increased downtime, or, in some cases,
extended, unplanned shutdowns. at
this time, it is not economically viable to
design, build and operate a plant that
produces one or more rare gases and
34
100
90
80
70
60
50
40
30
20
10
Xenon
Krypton
1
2
3
4
5
Liquid flowrate, percent of oxygen production
Figure 2. The graph shows the impact of O2 withdrawal
rate on the recovery recovery of Xe and Kr from the
reboiler sump
vents the oxygen, nitrogen and argon.
there is potential to recover rare gases
from other chemical processes that use
large volumes of air, like some ammonia
processes, but it has not proven to be
economically viable yet.
Neon production
removing non-condensable gases is
important to the design of an asu. one
of the key components of an asu is the
main condenser that operates as the reboiler
at the bottom of the low-pressure
column and the condenser at the top
of the high-pressure column. minimizing
the temperature difference between
the top of the high-pressure column and
bottom of the low-pressure column has
a direct impact on the air pressure required
to drive the process, as well as its
separation power. in some processes,
a non-condensable-gas vent is used to
remove the non-condensables before
they build up to unacceptable levels. it
is desirable to remove as little nitrogen
gas along with the non-condensable
gases as is practical, because removing
nitrogen-rich gas from the top of the
high-pressure column decreases the
liquid nitrogen reflux sent to the top of
the low-pressure column. oxygen and
argon recoveries typically improve as liquid
nitrogen reflux is increased. process
cycle selection has a huge impact on ne
recovery potential. in some processes,
significant fractions of the feed air may
bypass the high-pressure column, reducing
the ne available for recovery.
some processes produce refrigeration
by expanding a portion (nominally 10%)
of the vapor air feed directly to the lowpressure
column.
many modern processes send a significant
fraction of the air feed as liquid
ChemiCal engineering www.Chemengonline.Com June 2020
Recovery, percent
http://www.Chemengonline.Com
Chemical Engineering June 2020

Table of Contents for the Digital Edition of Chemical Engineering June 2020
