Miller et al.
Iron(II) chloride tetrahydrate (FeCl2‚4H2O) and iron(II) sulfate
heptahydrate (FeSO4‚7H2O) were obtained from Aldrich. Reagent
grade methanol, ethanol, and acetonitrile were deoxygenated with
an argon purge before being brought into the glovebox. The
anhydrous Na2SO4 (Aldrich) was used as received. CO was obtained
from Linde.
soluble. Water is an ideal solvent for such a process.
However, the known iron-diphosphine complexes are water
insoluble, and it is necessary to modify them accordingly.
The method we chose to impart water solubility was to use
phosphines that are water soluble. Sulfonated phosphines are
often used in this capacity,3,4 but we found that sulfonate
groups are often non-innocent and can affect the ability of
the iron compounds to react with N2 or other small
molecules. For example, when sulfonated phosphines were
used for the chemistry reported herein, N2 binding was
greatly decreased due to competitive binding of the sulfonate
group to the iron center.4d To avoid this problem, we decided
to focus on phosphine ligands that have hydroxyl groups as
the water-solubilizing entity. Examples of the phosphines
that have been developed are the 1,2-bis(bis(hydroxyalkyl)-
phosphino)ethane ligands and the related methoxy derivative,
shown below.4
Instrumentation and Procedures. 31P{1H} NMR were run on
either a VARIAN GEMINI 2000 NMR spectrometer (at BRI) at
121.47 MHz and referenced externally to 1% H3PO4 or on a Varian
Unity/Inova 300 spectrometer (at UO) at an operating frequency
of 299.95 and 121.42 MHz for 1H and 31P nuclei, respectively. The
1H and 31P{1H} NMR obtained on the latter instrument were
referenced to the solvent peak and to an external standard of 1%
H3PO4 in D2O, respectively. The samples were sealed under argon
in 5 mm tubes fitted with Teflon valves. Infrared spectra were
recorded on a Perkin-Elmer Paragon 1000 FT infrared spectropho-
tometer (at BRI) or on a Nicolet Magna 550 FT-IR spectrometer
(at UO) with OMNIC software. Samples were prepared as either
Nujol mulls with NaCl or AgCl windows or in solution with CaF2
or ZnSe cells. UV-vis spectra were recorded with a Perkin-Elmer
Lambda 6 spectrophotometer. Elemental analyses were performed
by E+R Microanalytical Laboratory, Inc., Corona, NY.
Kinetics Studies. All of the kinetics experiments were performed
in H2O in a thermostated cell. The water was purified to a resistivity
of 17-18 MΩ‚cm with a Barnstead Nanopure II system. The water
was bubbled with oxygen-free N2 for 1 h prior to use. The reactions
were monitored by recording the disappearance of cis-Fe(DHBuPE)2-
(SO4) at λmax ) 500 nm. The kinetics experiments in the presence
of Na2SO4 were carried out at 293 K. Values of the rate constants
reported are an average of at least three individual measurements.
In this paper, we report the synthesis and characterization
of the complexes formed in the reactions of these ligands
with FeCl2‚4H2O, and we compare these compounds to the
analogous complexes that have water-insoluble chelating
phosphine ligands such as depe (1,2-bis(diethylphos-
phino)ethane). In addition, we report the reactions of FeSO4‚
7H2O with the 1,2-bis(bis(hydroxyalkyl)phosphino)ethane
ligands. The reactions of these complexes with N2, CO, and
CH3CN are also discussed, and the crystal structures of the
cis-Fe(DHPrPE)2SO4, trans-Fe(DHPrPE)2(CO)SO4, trans-Fe-
(DHBuPE)2Cl2, trans-[Fe(DHBuPE)2(CO)(Cl)][B(C6H5)4],
trans-Fe(DMeOPrPE)2Cl2, trans-Fe(DMeOPrPE)2Br2, and
trans-[Fe(DHBuPE)2Cl2]Cl complexes are reported. As part
of this investigation, we also studied the effect of solvent
and counterion on the products obtained in the preparatory
reactions. These investigations led to the unexpected and
previously unreported results that the transition metal
compound stereochemistry is dependent on solvent, coun-
teranion, and alkyl chain length on the phosphine.
X-ray Structural Analyses. Crystals of 1, 5, 8, 9, and 10 for
X-ray work were manipulated under hydrocarbon grease and sealed
in special glass capillaries in the glovebox. The data crystals of 2
and 4, which remained stable during data collection, were coated
with epoxy resin and mounted on glass fibers. Cell dimensions and
orientation matrices were determined from the setting angles of an
Enraf-Nonius CAD-4 diffractometer for 25 centered reflections in
the following θ ranges: 1, 15-16°; 8, 14-15°; 2, 13-14°; 9, 12-
14°; 4, 14-15°; 5, 14-15°; and 10, 14-15°. Table 1 contains a
summary of crystal data and the final residuals; fuller tables with
particulars of data collection and structure refinement are in the
Supporting Information. Data were collected to θ 25° for 1, 2, 4,
5, and 10 and to θ 22.5° for 8. Because of crystal decay, data for
the weakly diffracting 9 were collected from two crystals, ranges
1.5-20° and 20-23° θ, and the data sets were corrected for decay
before they were combined. Structure solutions were obtained from
SIR92 E-maps.6 Small absorption corrections were applied to the
Experimental Section
Materials and Reagents. Unless otherwise noted, all manipula-
tions were carried out in an argon-filled Vacuum Atmospheres Co.
glovebox or on a Schlenk line with nitrogen. The 1,2-bis(bis-
(hydroxyalkyl)phosphino)ethane ligands were prepared as reported
previously.4b 1,2-Bis(diethylphosphino)ethane (depe) was obtained
from Strem Chemical Co. and used without further purification.
(5) The 1H NMR spectra of the molecules reported herein were relatively
broad and generally similar to the spectra of the uncoordinated
phosphine ligand. Therefore, they were nondiagnostic in terms of
checking for either purity of the sample or the identity of the molecule.
In contrast, the 31P{1H} NMR spectra of the molecules were simple,
had no overlapping peaks, and were spread over a wide range of
chemical shifts. Thus, they provided an excellent means for character-
izing and identifying the products and for checking their purity. For
(2) Lyon, D. K. Fe Phosphine Complexes for N2 Removal from Natural
Gas, U.S. Patent 5 225 174, 1993.
(3) See, for example: Cornils, B.; Wiebus, G. ChemTech 1995, January,
33.
(4) (a) Nieckarz, G. F.; Weakley, T. J. R.; Miller, W. K.; Miller, B. E.;
Lyon, D. K.; Tyler, D. R. Inorg. Chem. 1996, 35, 1721-1724. (b)
Baxley, G. T.; Miller, W. K.; Lyon, D. K.; Miller, B. E.; Nieckarz,
G. F.; Weakley, T. J. R.; Tyler, D. R. Inorg. Chem. 1996, 35, 6688-
6693. (c) Baxley, G. T.; Weakley, T. J. R.; Miller, W. K.; Lyon, D.
K.; Tyler, D. R. J. Mol. Catal. A 1997, 116, 191-198. (d) Miller, W.
K.; Lyon, D. K.; Tyler, D. R. Unpublished observations.
1
the sake of completeness, the H NMR spectra of the complexes are
1
reported in the Experimental Section. For comparison, the H NMR
spectra of the ligands are as follows: DHPrPE (MeOH), 1.52 (m, br,
8 H, 4 × P-CH2), 1.57 (m, br, 4 H, P-CH2CH2-P), 1.64 (m, br, 8
H, 4 × CH2), 3.61 (t, 8 H, 4 × CH2-OH), 4.8 (s, OH); DHBuPE
(D2O), 1.47 (m, br, 20 H), 1.59 (m, br, 8 H), 3.54 (t, 8 H, 4 × CH2-
OH), 4.80 (s, OH); DHPePE (MeOH), 1.38 (m, br, 28 H), 1.47 (m,
br, 8 H), 3.46 (t, 8 H, 4 × CH2OH), 4.8 (OH).
(6) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Cuagliardi, A.; Burla,
M. C.; Polidori, G.; Camalli, N. J. Appl. Crystallogr. 1994, 27, 435.
5454 Inorganic Chemistry, Vol. 41, No. 21, 2002