Solubility of Ferrocene in Organic Solvents
Walter R. May
SFA International, Inc.
Houston, TX
Of all metals with catalytic effects on the combustion of hydrocarbons, iron is one of the
best. In the literature, iron reduces particulate matter in boiler and engine exhausts
reaching as high as 50%.1 SFA International’s patented combustion catalyst technology
involves the use of iron combined with magnesium.2 One of the requirements of a fuel
additive is that metal compounds in the additive must be soluble in the fuel.
Ferrocene contains 30% iron by weight and can be used as a source of iron in fuel
additives. Ferrocene is sold as a fuel additive in liquid and solid forms. In solid form, it is
available as “crumbs” and “caplets.” The advantage of using the solid form of ferrocene
is that it is easy to ship and add to a fuel tank. Unfortunately, users of solid forms of
ferrocene as a combustion catalyst have reported a wide range of results that are
mostly negative.3 The purpose of this paper is to evaluate the chemistry of ferrocene
and solubility in organic solvents to find the reasons for generally negative results with
solid forms of ferrocene as a combustion catalyst.
Ferrocene
Sandwich compounds and ferrocene have been known since 1951 when Pauson and
Kealy reacted cyclopentadienyl magnesium bromide Grignard Reagent with ferric
chloride. They obtained a light orange powder of “remarkable stability.”4 It is stable to
400o without appreciable decomposition.5 Professor Mark M. Jones of Vanderbilt
University (the author’s Ph.D. advisor) wrote extensively on ferrocene in his book
Elementary Coordination Chemistry.6
1
Boiler Fuel Additives for Pollution Reduction and Energy Saving, Ed. R. C. Eliot, Noyes Data Corporation,
Park Ridge, NJ, 1978.
2
See SFA’s web site at www.SFAInternational.com for more information.
3
Sources for this statement are personal and confidential conversations with several people who have
had experience with using solid forms of ferrocene as a combustion catalyst. Results have been sporadic
and overwhelmingly negative.
4
http://en.wikipedia.org/wiki/Ferrocene
5
Inorganic Reactions and Structure, Edwin S. Gould, Henry Holt and Company, New York, NY., 1955.
6
Elementary Coordination Chemistry, Mark M. Jones, Prentice-Hall, Inc., Englewood Cliffs, NJ, 1964.
1
According to the Handbook of Chemistry and Physics,7 ferrocene has a molecular
weight of 186.04, 249o b.p. and 172.5o m.p. It is notable that there are no data
regarding solubility of ferrocene in polar or non-polar solvents. Ferrocene is 30.02%
iron by weight.
Ferrocene - Dicyclopentadienyl Iron(0)
Ferrocene as a Fuel Additive
While ferrocene is made only in China at this time, there are a few importers of the
product into the U.S. that hold Tier One testing and EPA Registration for on-road use.
Most of these products are the solid form of ferrocene. The cost of Tier One
registration testing is about $300,000 in an EPA acceptable laboratory. A number of
companies have EPA registered re-branded products purchased from the holders of the
original Tier One registration. Therefore, ferrocene has been very attractive as a lowcost on-road EPA Registered fuel catalyst with a minimum of problems shipping and
applying solid forms.
Metallic fuel additives can only be effective if dissolved at the molecular level in the fuel
or, at a minimum, dispersed at sub-micrometer size levels. SFA’s FuelSpec® 116 and 117
products are oil-soluble iron carboxylates (salts of organic acids) that are completely
soluble in aliphatic hydrocarbon fuels. SFA’s FuelSpec® 118 and 119 catalysts are based
on an iron colloidal dispersion. FuelSpec® 118-1502 has a median particle size of 0.007
micrometers.
To be an effective combustion catalyst, ferrocene must also be soluble in naphtha
(gasoline) and Diesel fuels. The Wikipedia article states that ferrocene is “soluble in
most organic solvents.” Siddiqi and Atakan8 studied the dynamic equilibrium between
gas and liquid phases. They found ferrocene to be “only weakly soluble in these
solvents” which included ethanol, toluene, hexane and 2,2,4-trimethylpentane
(isooctane). With this minimal information, we undertook an investigation of the
solubility characteristics of solid forms of ferrocene in organic solvents to find out if this
is the cause of observed anomalies in performance of the product as a combustion
catalyst.
7
8
Handbook of Chemistry and Physics, Ed. Robert C. Weast, CRC Press, Inc., Boca Raton, Florida, 1984-85.
Siddiqi, A., Atakan, B., J. Chem. Eng. Data, 2006, 51 (3), pp. 1092-1096.
2
Solubility of Ferrocene in Organic Solvents
A study was carried out at the University of North Texas on the “Solubility of Ferrocene
in Organic Nonelectrolyte Solvents, Comparison of Observed versus Predicted Values
Based upon Mobile Order Theory.”9 This paper includes experimental data on ferrocene
in forty-six organic solvents compared with theoretical calculations for solubility based
on mobile order theory. The solvents included a number of aliphatic compounds,
aromatic compounds including benzene, toluene and xylene, alcohols, esters, ketones
and organic sulfur compounds. The experimental data are presented in Table I.
Solubility ranged from a high of 5.59% Fe for benzene to 0.09% for 1-Butanol. Xylene
isomers had relative high solubility’s in the 3.5 to 4.0% range. Aliphatic solvents
dropped off to the 1.0 to 1.5% ranges. Chlorinated hydrocarbons were high as well as
carbon disulfide and pyridine. This indicates that solvents with higher levels of
aromaticity were better solvents than non-polar aliphatic solvents. Oxygenated
compounds were the poorest solvents.
An examination of the structure of ferrocene (given on page two) indicates some
aromaticity from the two sets of double bonded carbons in each cyclopentadienyl ring.
However, those pi bonds are directed toward the strongly Lewis acid iron atom between
the rings. The compound has weak aromatic character and a lack of non-polar areas
commensurate with aliphatic properties. There are no polar areas for attraction to
electron acceptors or donors. This eliminates oxygenated solvents. There is no reason
to expect ferrocene to be especially soluble in any solvent, polar or non-polar.
Ferrocene has no solubility in water and inorganic electrolyte solutions.
SFA Solubility Results
SFA recently undertook an evaluation of ferrocene solubility in a highly aromatic
naphtha solvent and ultra low sulfur No. 2 (Diesel) fuel. We found a slow dissolution
rate for ferrocene at 1% iron level (3.33% ferrocene) in heavy aromatic naphtha and an
even slower dissolution rate in Diesel fuel. We were able to dissolve 2% iron (6.66%
ferrocene) in heavy aromatic naphtha with vigorous mixing over about 3 days. We
attribute this to the fact that aromatic fuels are better solvents for ferrocene than highly
aliphatic fuels.
Ferrocene was added to dyed off-road LED fuel at a 6.66% concentration to yield 2%
iron in the solution. The product was vigorously shaken one time per day over a
thirteen day period. The following time-dated photographs show the results. Figure 1 is
a picture of the sample immediately following addition of the ferrocene to LED and
9
DeFina, K. M., Ezell, C., Acree, Jr., W. E., Physics and Chemistry of Liquids, 39, (6) Nov. 2001, Pages 699710.
3
vigorous shaking. Figure 2 shows un-dissolved ferrocene twelve days later. Figure 3
shows material clinging to the bottom of the bottle after it was inverted on the 13th day.
Our work with ferrocene indicates that to dissolve the material in a solvent at percent
levels, a highly aromatic solvent must be used. Vigorous agitation plus heating will help
dissolution.
It should be pointed out that iron is generally added to a fuel at 5 to 50 ppm whereas
the samples evaluated in this study were at 1% and 2% or 10,000 and 20,000 ppm iron.
Ferrocene would be expected to dissolve in aliphatic fuels at additive dosage levels with
time and vigorous agitation.
There are three things to consider with the use of solid forms of ferrocene as a fuel
additive.
1. Because of low solubility and slow dissolution rates of ferrocene in typical LED
aliphatic fuels, addition to a tank with minimal mixing is almost certain to result
in poor results. Although ferrocene will dissolve with time in a static system due
to thermal convection, localized concentration in the area of the solid material
exceeding solubility limits will result in slowing or stopping dissolution of
ferrocene in the fuel. Solid forms of ferrocene should never be put in fuel
without thorough mixing which is generally not available in small contained
tanks in vehicles and other equipment.
2. “Caplets” have a much smaller surface area to weight ratio than “crumbs”
greatly reducing rate of dissolution. “Caplets” and large compressed forms of
ferrocene should never be used and sales of these products should cease.
3. All ASTM No. 2 distilled fuels including LED fuels contain water that settles from
the fuel and accumulates on the bottom of fuel tanks. Solid ferrocene has a
higher density than fuel and, when added to a tank, immediately settles to the
bottom of the tank in the water phase. Ferrocene is insoluble in water and this
material will never dissolve. An eventual result can be plugging fuel filters.
Ferrocene Solutions
A limiting factor on the use of dissolved ferrocene as a fuel additive is the maximum
concentration of iron that can be achieved in these solutions. The practical limit is 2%
iron in highly aromatic solvents. To achieve 10 ppm iron in fuel with a 2% iron solution,
a dosage rate of one part additive to 2,000 parts fuel by weight is required.
SFA International has developed iron products with up to 9% iron in an oil soluble form
and 18% iron in a colloidal dispersion for industrial applications. The resulting lower
dosage rates yield much higher flexibility in application of the product and more
4
economically effective results. SFA discovered a synergistic combination of iron and
magnesium that is a superior combustion catalyst to iron alone.
In boiler and
combustion turbine exhaust measurements, SFA’s iron-magnesium catalyst yields 90%
reduction in particulate matter in exhausts10 compared with 50% expected from the
literature.1
A study was carried out on a railroad locomotive using kerosene fuel with about 3,000
ppm sulfur that yielded 10% in switcher duty and 5% in road duty improvement with a
ferrocene solution catalyst.11 SFA measured 12.0% fuel savings with the same fuel in
600 HP switch engine and 12.8% savings in a 1,500 HP road engine.12 It is estimated that
SFA’s iron-magnesium catalyst yields 50% better fuel economy than iron alone.
Conclusion
Sporadic results from the use of ferrocene as a combustion catalyst can be explained by
observations on the solubility of ferrocene in organic solvents. Our conclusion from this
evaluation is that solid forms of ferrocene such as crumbs should be avoided. While
crumbs can dissolve under ideal conditions, opportunities for the material to remain in
solid form remain. Caplets with even poorer dissolution rates should be avoided and
sales of these products should be stopped.
Ferrocene can be used in combustion catalyst or fuel additive formulations when
dissolved in a solvent so that it will disperse thoroughly into the fuel. There are better
approaches to adding iron to a fuel exemplified by SFA’s higher concentration product
line. Without mitigating circumstances such as EPA Registration requirements for onroad use, the user would be well advised to use SFA International’s much higher
performing iron and magnesium combinations.
Performance of a combustion catalyst containing iron (or iron plus magnesium) requires
sufficient dosage. Optimum dosage levels are 10 ppm iron in compression-ignited
reciprocating engines and 50 ppm iron in continuous firing equipment such as
combustion turbines, boilers and process heaters. Lower dosage rates based on
economic considerations are ill advised and will not produce desired results.
10
May, W and Hirs, E., “Catalyst for Improving the Combustion Efficiency of Petroleum Fuels in Diesel
th
Engines,” 11 Diesel Engine Emissions reduction Conference, August 21-25, 2005, Chicago, IL.
11
Note the 100% variation in the two measurements that could be attributed to mixing in the fuel. These
types of observations are common with ferrocene based catalysts.
12
Verification Report was prepared by Greenhouse Gas Technology Center, Southern Research Institute,
Research Triangle Park, NC, for EnviroFuels, Inc. The test was run on a St. Lawrence and Atlantic Railroad
(Division of Genese & Wyoming) 3,000 HP EMD GP-40-3 locomotive. This test was run in 2005 in the state
of Maine. Although an analysis of the fuel used in this test was not given in the report, it is estimated that
the fuel was similar to that used in Lincoln and Plymouth Railroad tests with SFA’s iron-magnesium
catalyst in new Hampshire in the same year. The complete test is available at www.envirofuelsllc.com.
5
Figure 1. Immediately following addition of ferrocene to off-road dyed Diesel fuel.
12-Apr-09.
6
Figure 2. Undissolved ferrocene after 12 days.
24-Apr-09
7
Figure 3. Undissolved ferrocene shown sticking to bottom of bottle.
25-Apr-09
8
Table I. Solubility of Ferrocene in Organic Non-Electrolyte Solvents
Solvent
n-Hexane
n-Heptane
n-Octane
n-Nonane
n-Decane
n-Hexadecane
Cyclohexane
Methylcyclohexane
Cyclooctane
2,2,4-Trimethylpentane
t-Butylcyclohexane
Benzene
Toluene
Ethylbenzene
o-Xylene
m-Xylene
p-Xylene
Dibutyl ether
Methyl t-butyl ether
1,4-Dioxane
Methanol
Ethanol
1-Propanol
2-Propanol
1-Butanol
2-Butanol
2-Methyl-1-propanol
2-Methyl-2-propanol
1-Pentanol
2-Pentanol
3-Methyl-1-butanol
2-Methyl-2-butanol
1-Hexanol
2-Methyl-1-pentanol
4-Methyl-1-pentanol
1-Heptanol
1-Octanol
2-Ethyl-1-hexanol
1-Decanol
2-Ethyl-1-hexanol
Solvent
Mol. Wt.
86.178
100.205
114.232
128.259
142.286
226.448
98.189
113.224
109.192
114.232
154.297
78.114
92.141
106.168
106.168
106.168
106.168
130.232
88.151
88.108
32.043
46.070
60.097
60.097
74.124
74.124
74.124
74.124
88.151
88.151
88.151
88.151
102.178
102.178
102.178
116.205
130.232
130.232
158.286
130.232
Mole Fraction
Solubility
0.022600
0.024890
0.027130
0.029010
0.030970
0.039630
0.033000
0.033720
0.046800
0.021790
0.036120
0.087586
0.083210
0.077030
0.080140
0.074360
0.077850
0.051070
0.041200
0.068300
0.003298
0.005976
0.008917
0.007078
0.001181
0.010270
0.009621
0.009215
0.013520
0.012630
0.012250
0.015540
0.017350
0.014260
0.013430
0.020500
0.022150
0.016670
0.027670
0.016670
% Sol.
4.75%
4.52%
4.34%
4.15%
4.01%
3.28%
6.07%
5.42%
7.72%
3.50%
4.32%
18.61%
15.49%
12.76%
13.24%
12.34%
12.89%
7.14%
8.31%
13.40%
1.88%
2.37%
2.71%
2.16%
0.30%
2.54%
2.38%
2.28%
2.81%
2.63%
2.55%
3.22%
3.11%
2.57%
2.42%
3.24%
3.13%
2.36%
3.24%
2.36%
% Fe
1.43%
1.36%
1.30%
1.25%
1.20%
0.98%
1.82%
1.63%
2.32%
1.05%
1.30%
5.59%
4.65%
3.83%
3.98%
3.70%
3.87%
2.14%
2.50%
4.02%
0.57%
0.71%
0.81%
0.65%
0.09%
0.76%
0.71%
0.68%
0.84%
0.79%
0.77%
0.97%
0.94%
0.77%
0.73%
0.97%
0.94%
0.71%
0.97%
0.71%
Table I Continued
Solvent
1-Decanol
Cyclopentanol
Butyl acetate
Ethyl acetate
Methyl acetate
2,3-Dichloroethane
1-Chlorobutane
1-Chlorooctane
Tetrachloromethane
2-Propanone
Acetonitrile
Dimethyl sulfoxide
Carbon disulfide
Pyridine
Solvent
Mol. Wt.
158.286
86.135
116.162
88.108
74.081
98.960
92.569
148.677
153.823
95.550
41.050
78.130
76.140
79.100
Mole Fraction
Solubility
0.027670
0.017740
0.055800
0.043000
0.032580
0.077350
0.059620
0.060620
0.06920
0.02400
0.00756
0.01410
0.06690
0.07048
% Sol.
3.24%
3.75%
8.65%
8.67%
7.80%
13.61%
11.30%
7.47%
8.25%
4.57%
3.34%
3.29%
14.91%
15.13%
% Fe
0.97%
1.13%
2.60%
2.60%
2.34%
4.09%
3.39%
2.24%
2.48%
1.37%
1.00%
0.99%
4.48%
4.54%