Re: Inert Gas in Fuel Tanks

From:         Bob Standaert <>
Organization: Texas A&M University
Date:         08 Nov 96 05:24:22 
References:   1 2
Followups:    1 2
Next article
View raw article
  or MIME structure

Several of the comments on inert gases caught my attention, so I thought
I'd offer up some tidbits.  Even though the point is moot, there are
some interesting things to think about.

Nitrogen gas has a density of 1.25 g/l at 0 C, 1 atmosphere pressure.
For the liquid, it is 808 g/l at -196 C, for a volume ratio of 646:1.
If the fuel capacity of a 747 is about 200 kl, you would need 250 kg, or
310 l, of lN2, to replace it.  The dewars (vacuum jacketed storage
vessels) we have hold 200 l of lN2 and weigh about 120 kg empty.  Two
full tanks on a plane adds up to about 560 kg plus plumbing and
hardware, and the cost for the liquid would be about $40 at our price.
(Perhaps the airlines could make up for it by charging the passengers
extra for the spectacle of the refill -- lots of fog and noise coming
from the belly of the plane!)

Nitrogen boils at -196 C and cannot be liquefied above its critical temp
of -147 C.  The liquid is therefore kept and handled in insulated
containers at ambient pressure or a bit above; our large stationary tank
(about 40,000 l) is kept at 100-150 PSI, and the 200 l-dewars are rated
at 200 PSI).  A failure in the overpressure protection would lead to
rupture of the tank, and smaller dewars are usually open to the
atmosphere.  We use the stuff as the liquid and for the most part don't
want it to evaporate.  If you want the gas, you need to heat the liquid.

High pressure gas would probably present too much of a weight problem.
A big, steel lab cylinder (10 x 55") holds about 8500 l (gas at ambient
pressure), weighs in at around 75 kg, and is pressurized to 2500 PSI.
It seems to take a lot of steel to contain that pressure, even for a
small-diameter tank.  Aluminum might save you a third to a half, but the
tanks are still heavy.  You can liquefy CO2 at a modest  pressure
(around 1000 PSI) and stuff about twice as much (in terms of net gas
volume) into the same tanks.  If the head is sheared off any high
pressure gas cylinder, as happens occasionally when one is mishandled,
it becomes a missile that can penetrate concrete block walls, and
presumably aircraft hulls.

Weight and headaches aside, liquid nitrogen is not without its hazards.
Some things you need to worry about from a leak:

1.  Oxygen displacement.  A big leak that evaporated in a confined area
would asphyxiate any living being within (what is the volume of the
cargo hold?).

2.  Cold damage.  Cooling things to -200 C tends to change their
mechanical properties.  The classic demonstration is the tennis ball or
rubber band (or, in the old days, goldfish) cooled in lN2 that shatters
into bits when struck.  Severe frostbite develops quickly on skin
exposed to lN2.  Plastics crack, adhesives fail, and metals become more
brittle.  Thermal contraction, water condensation, or ice formation
could cause a mechanical malfunction, and a big leak of liquid into the
fuel tank would give you kerosene gummi-bears.  You get the idea.

3.  Oxygen condensation.  Surfaces cooled by lN2 can condense O2 from
the air; lO2 has the nasty property of reacting with any combustible
material it touches, often explosively.  In the lab, this problem arises
most commonly when air gets into a liquid-nitrogen cooled vapor trap.

This all leads up to a point that's already been made -- an lN2 system
would be a headache.  On-the-fly deoxygenation of air may be feasible,
but I am not an expert in the area of gas separations.  My impression is
the technology has come a long way in the last 20 years.  Reliable,
noncryogenic, small scale separators (20,000 l/hr) capable of generating
N2 of 95+% purity are available for on-the-ground installations.
However, I have no idea how big they are, how much they cost, or how
much power they consume.  One commonly used separation method is to pass
compressed air (50-500 PSI) over selectively permeable, hollow-fiber
polymeric membranes.  Oxygen and water permeate to the inside of the
fibers, and the nitrogen (78% of air) just sails on by.  The oxygen can
be recovered, too.  Nice systems, and no moving parts outside the
compressor.  I'll leave it to others to comment on the practicality of a
small mechanical or ram-air compressor to do the job.  As for
centrifugal gas separation in the engine, which one post mentioned, the
only thing spun out of this scenario is a yarn.

Regards to all,

Bob Standaert
Assistant Professor of Chemistry
Texas A&M University