gruns (Stranger)
02-15-02 06:44
No 269996
      Iron pentacarbonyl  Bookmark   

Howdy all, this is my first post, I hope I don't screw it up.
Let me begin by saying that I am not a chemist, in fact I know
next to nothing about chemistry.  I do, however, know how to
gather and collate information, and this is what I'm basing
this post from.

Anyway, I noticed in Rhodium's isomerization article the bit
about using small amounts (~1g/200g olefin) of Iron Pentacarbonyl
to isomerize olefins.  I looked up that Fe(CO)5 and found the
It's a spectacularly poisonous (causes cyanosis) viscous yellow
liquid with an autoignition temp of 50C and a BP of 103C, if
decomposed somewhere around 250C it leaves pure iron particles.
It is produced commercially by exposing iron to carbon monoxide
at 120C under a pressurized, inert atmosphere.

Here's what I'm proposing, a method to isomerize small amounts
(less than 200g) of olefin in good yield.

*Gruns checks his notes*

HCOONa  --heat--> CO + NaOH
(Alternately, formic acid can be added to sulfuric for the
same effect, although there'd probably be a little water vapor
in there somewhere.)

Fe + 5CO --heat--> Fe(CO)5

(280mg Fe + 720mg CO --heat--> 1g Fe(CO)5)

CO occupies 862ml/g at room temp/pressure (or so said a commercial
supplier of CO, well, actually he said it was .862m3/kg[1]), so
the functional limit for each reaction would be the size of
the reaction vessel unless you have access to compressed CO.

As I said previously, I am not a chemist, I do not yet know the
method by which it works [not for lack of trying, I couldn't find
a patent for this and I have yet to locate that chemistry journal
mentioned in the isomerization text], so the data I work from is
the empirical data from the text.


Under an inert atmosphere charge a 1l vessel capable of being
pressurized to 30psi with CO (having a spgrav of .9676),
throw in 280mg of iron powder and cap with a stopper
allowing for later pressurization.  (I realize this is an
excess of CO, but as "Bob" always says, "Too much is always
better than not enough.")

On that, I'm sure that somewhere out there exists a stopper
made just for this, but for the purposes of the ghetto, it
seems like a one-holed stopper fitted with a bicycle-tube
valve stem and held in place with some champagne cork wiring
would do the trick.  A champagne bottle would probably work
well, I don't know what pressures a RBF is capable of handling.

In any case, pressurize the container to "1 or 2 bars"[2], or
15-30psi with the inert gas and gently raise the temp to 120C
with an oil bath.

Once the conversion is complete, remove from heat and chill to
as cold as you can goddamned get it, 'cos Fe(CO)5 is some NASTY
STUFF, breathing the vapors can kill you, or so says every
MSDS I've read concerning this chemical.  Once cold, carefully
depressurize, decork, add reagents and proceed with the
isomerization as detailed by the aforementioned document,
adjusted to use 1g of Fe(CO)5.

Considerations, questions, requests:

Would unreacted Fe affect the isomerization reaction?
Would the CO react with the brass in the valve?
How long would the isomerization take?
Autoignition is only a concern if there's oxygen to burn with...right?
Would CO2 work as the inert atmosphere? Nitrogen?
Could someone post the full text of the ref in the isomerization text?
Is Fe(CO)5 too dangerous to work with, even with the precautions I've

Okay, this is total speculation on my part, but, sodium formate
decomposes into NaOH and CO upon heating, so, what if one were
to place both iron powder and sodium formate into a vessel also
containing an inert atmosphere, pressurize... maybe not mix
them, but put them in different areas on the bottom of the vessel
to heat separately, then the remaining NaOH would be there, ready
to assist in the isomerization of the olefin.  Unless of course,
liquid Fe(CO)5 reacts with NaOH, in which case that wouldn't
work.  Fe(CO)5 has a lower vapor density than carbon dioxide.

[1] -
[2] -

Pertinent text of latter link, a site discussing the chemistry of using
ores and such found on mars as construction materials/precursors for
"ordinary" substances.

Iron Refinement
3Fe2O3 + 9CO +> 6Fe + 9CO2 leaving metallic Iron
(((3Fe2O3 + CO +>2Fe3O4 + CO2)))
(((Fe3O4 + CO +> 3FeO + CO2)))
(((FeO + CO +> Fe + CO2)))
Leaving metallic Iron would require some form of electromagnetic
separator or
rake to remove the Fe from the regolith.

However, using Iron Carbonyl production, the Iron is drawn off as a
carbonyl vapor. So actually, we do not need an electromagnet...
unrefined regolith goes in (maybe sorted by particle size) the mix
is heated to 120 C with CO and the pressure is raised a bar or two.
The gas is drawn off leaving the SiO2, CaO, etc all as they were.
The gas is depressurized and cooled with the Fe(CO)5 condensing out
for our use!!
Iron Carbonyl Production
Fe + 5CO => Fe(CO)5 120 C liquid, vaporizes at modest pressures


Bonus time.

US Patent: 3,091,641

While searching patents for iron pentacarbonyl, I came across a number
of them using it as a catalyst for the production of tertiary amines
from aliphatic ketones.


I actually found a number of these, most of them under the unassuming
title of "preparation of tertiary amines" and using a combination of
ruthenium/rhodium/osmium etc carbonyls with iron pentacarbonyl as
catalysts, but one of them listed only Fe(CO)5 and CO aside from
the usual suspects.  If indeed Fe(CO)5 is found safe and effective for
the symptomatic treatment of 1-alkenes, it may well turn out to be a
big pain in the ass for the suits if the same is true for this reaction.

Here's some text to get you wet:

The present invention relates o the preparation of tertiary amines and,
more particularly, to the preparation of tertiary amines from a
secondary amineand an aliphatic ketone by reaction with carbon monoxide
and water in the presence of an iron carbonyl catalyst.  The reaction
can be represented by the following equation:
  H         0 
  |         ||              R     R"
R-N-R' + R"-C-R'"+ CO --->   \   /
                              N-CH + CO2    
                             /   \
                            R'    R'"
wherein R, R', R", and R'" each represents an alkyl group or an aralkyl
group or a cycloalkyl group, the alkyl portion or group containing from
1 to 7 carbon atoms and wherein at least one of the R" and R'" groups
is an alkyl group.  It is preferred that the groups R, R', R", R'"
represent lower alkyl groups containing from 1 to 4 carbon atoms.

Unfortunately I can't post the entire text of the patent as it is only
available in TIFF format and I don't have an OCR app handy.

Please, please, rip this document apart.