Solubility of trans-Co2(CO)6 [3,5-bis(CF 3)C6H3P(i-C3H7) 2] in dense carbon dioxide

Article abstract of DOI:10.1016/j.jorganchem.2004.11.053

The title compound (1) was prepared by the reaction of 3,5-bis(CF ₃)C₆H₃P(i-C₃H₇) ₂ (L1) and Co₂(CO)₈. Its solubility in supercritical carbon dioxide was measured at varying temperatures and pressures using a modified analytical extraction device. Solubility data were determined in the temperature and pressure ranges between 40 and 70°C and between 100 and 300 bar, respectively. The solubility of 1 is lower compared to (p-CF ₃C₆H₄)₃P, but much higher than for transition metal complexes bearing phosphines without fluorinated substituents.

Full text of DOI:10.1016/j.jorganchem.2004.11.053

Journal of Organometallic Chemistry 690 (2005) 1467–1473
www.elsevier.com/locate/jorganchem
Solubility of trans-Co2(CO)6 [3,5-bis(CF3)C6H3P(i-C3H7)2] in
dense carbon dioxide
*
Nicolaus Dahmen, Pia Griesheimer, Piotr Makarczyk, Stephan Pitter , Olaf Walter
Forschungszentrum Karlsruhe, Institut fu¨r Technische Chemie, Bereich Chemisch-Physikalische Verfahren, P.O. Box 3640,
D-76021 Karlsruhe, Germany
Received 12 July 2004; accepted 18 November 2004
Available online 22 January 2005
Abstract
The title compound (1) was prepared by the reaction of 3,5-bis(CF3)C6H3P(i-C3H7)2 (L1) and Co2(CO)8. Its solubility in super-
critical carbon dioxide was measured at varying temperatures and pressures using a modified analytical extraction device. Solubility
data were determined in the temperature and pressure ranges between 40 and 70 ꢀC and between 100 and 300 bar, respectively. The
solubility of 1 is lower compared to (p-CF3C6H4)3P, but much higher than for transition metal complexes bearing phosphines with-
out fluorinated substituents.
ꢁ 2004 Elsevier B.V. All rights reserved.
Keywords: Cobalt carbonyl complexes; Supercritical carbon dioxide; Metal complex solubilities
1. Introduction
SCFs as functions of pressure have been reported by
several research groups and reviewed over the past years
[5,6].
Multi-phase catalysis allows for process steps, by
means of which the reacting components can be sepa-
rated easily for further purification or reprocessing of
the catalyst and/or the unreacted substrate. Apart from
aqueous systems and those based on ionic liquids, po-
tential use of supercritical fluids (SCFs) was particularly
investigated in the past decade [1–3]. SCFs in general al-
low to combine a homogeneous reaction in a single
phase with a subsequent separation step, e.g., in a
two-phase system only by adjusting temperature and/
or pressure. Additionally, the high compressibilities of
supercritical fluids in particular near their critical points
are supposed to result in pressure effects on reactions in
SCFs far more than at low pressures [4]. Investigations
on variations of selectivities or rates of reactions in
Our interest is directed towards the fundamental
understanding of which parameters adequately charac-
terize and influence a homogeneously catalyzed process
using supercritical carbon dioxide (sc CO2) as the
reaction medium (Tc = 31.05 ꢀC, pc = 73.9 bar). When
studying an integrated reaction-separation approach,
knowledge of the phase behaviour of the participating
components and especially of the catalyst solubilities,
on basis of which a later process development can take
place, obviously is indispensable. This knowledge allows
for the setting of optimum reaction conditions [7] as well
as for the design of an efficient post-reaction separation
[8,9].
Recently, we studied the solubility of different substi-
tuted phosphines in sc CO2, which may also act as co-
ligands in homogeneously catalyzed reactions [10,11].
It was found that fluorination generally allows for higher
solubilities compared to the unmodified derivatives.
*
Corresponding author. Tel.: +49 7247 822308; fax: +49 7247
822244.
E-mail address: stephan.pitter@itc-cpv.fzk.de (S. Pitter).
0022-328X/$ - see front matter ꢁ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.11.053

1468
N. Dahmen et al. / Journal of Organometallic Chemistry 690 (2005) 1467–1473
Based on empirical findings, similar fluorination con-
cepts have been also transferred to homogeneous transi-
tion metal catalysis [12].
derivatives of the trans-Co2(CO)6(phosphine)2 type, fur-
ther reaction of L1 in a benzene solution with Co2(CO)8
yields 1 [14]. The spectroscopic data of L1 and 1 listed in
Section 4.1 are as expected for these types of
compounds.
The present paper describes solubility measurements
of an organometallic transition metal pre-catalyst
trans-Co2(CO)6 [3,5-bis(CF3)C6H3P(i-C3H7)2] (1), refer-
ring to a process variant, in which product separation is
to be achieved by a thermally or pressure-controlled
phase separation, while the catalyst remains dissolved
in sc CO2. Complexes like 1 with electron-rich substitu-
ents at the phosphorous are promising candidates for
certain transition metal-catalyzed reactions, e.g., hyd-
roformylation of long-chained olefins. Most recently,
ethylene and propylene hydroformylation in sc CO2 uti-
lizing Co2(CO)6[P(4-CF3C6H4)3]2 was demonstrated by
Rathke and co-workers [13].
2.2. Molecular structure of the cobalt complex 1
Red crystals of 1 were obtained from a mixture of
dichloromethane and pentane. The molecular structure
of 1 is shown in Fig. 1.
1 crystallizes in the monoclinic space group P2(1)/c
with a centre of inversion between the cobalt atoms.
One of the CF3 groups is rotatory disordered (F31,
F32, F33). The phosphine ligands are orientated trans
to each other, which is the usual conformation of com-
plexes of the type Co2(CO)6(phosphine)2. All bond
lengths and angles are in the range of those of similar
derivatives with the cobalt centres being trigonal-bipyra-
midally coordinated and carbonyl ligands showing a
nearly perfect linearity [15–17].
2. Results and discussion
2.1. Synthesis of 3,5-bis(CF3)C6H3P(i-C3H7)2 (L1)
and trans-Co2(CO)6 [3,5-bis(CF3)C6H3P(i-C3H7)2]
(1)
2.3. Solubility measurements
L1 is synthesized by the Grignard reaction of 3,5-
(CF3)2C6H3MgBr with (i-C3H7)2PCl in Et2O with a
yield of 90%. In analogy with the syntheses of other
To determine solubilities of solids in sc CO2, an
extraction apparatus was constructed as shown in
Fig. 1. Molecular structure of 1 in the crystal. Selected bond distances (pm) and angles (ꢀ): Co(1)–C(1) 177.9(2), Co(1)–P(1) 219.63(6), Co(1)–Co(1A)
267.08(6), P(1)–C(10) 184.1(2), F(21)–C(17) 132.2(3), O(1)–C(1) 115.2(3) and C(1)–Co(1)–C(2) 118.87(11), C(1)–Co(1)–P(1) 96.77(7), C(1)–Co(1)–
Co(1A) 83.36(7), P(1)–Co(1)–Co(1A) 178.07(2), C(4)–P(1)–Co(1) 117.84(7), C(10)–P(1)–Co(1) 114.58(7).

N. Dahmen et al. / Journal of Organometallic Chemistry 690 (2005) 1467–1473
1469
Fig. 2 [18–21]. The apparatus mainly consists of a syr-
inge pump (A) (ISCO model 100DX), a supercritical
fluid extraction oven (C) containing an extraction ves-
sel of 10 ml volume (D) (ISCO SSX2-10), and a back-
pressure regulator (F) (JASCO model 880-81).
To allow for solubility measurements by a dynamic
procedure, equilibrium conditions have to be established
in the extraction cell. If a sufficiently low flow rate is ad-
justed, the CO2 passing the extraction cell is loaded with
an equilibrium substance amount in the steady state. To
determine the optimum flow rate of sc CO2, the solid is
extracted at different flow rates. Starting from a some-
what high value, decrease of the flow rate results in an
increased concentration of the solid in the fluid. At a
certain flow rate, the solution is saturated. Further
reduction of the flow rate does not cause any further
changes of the concentration.
and 31P NMR spectroscopy. According to Fig. 3, the
same complex quantities were extracted at flow rates
of 0.313 and 0.125 ml minꢀ1 within the experimental er-
ror of
5%. For this reason, a flow rate of
0.313 ml minꢀ1 was used for all further experiments.
The effect of temperature on the solubility of 1 was
investigated in a series of experiments at the same CO2
density (0.75 g cmꢀ3). The results are shown in Table
1. The temperature which generally may affect the solu-
bility of volatile compounds in compressed fluids has a
minor impact only on the solubility of 1 in the investi-
gated range. At temperatures of 50, 60, and 70 ꢀC,
approximately the same quantities of 1 are extracted.
The lower solubility at 40 ꢀC is believed to be mainly
caused by the temperature influence on the solubility
kinetics.
The solubilities of 1 at 50 ꢀC and various pressures
are compiled in Table 2. It is well known that one
important factor influencing the solubility in sc CO2 is
the density of the fluid [3]. Compared to the solubility
of different substituted phosphines in sc CO2 published
earlier, lower solubilities of the corresponding com-
plexes may be expected in general [22]. The findings re-
ported here confirm that 1, a metal complex coordinated
by fluorinated phosphine ligands, is less soluble than
2.4. Solubility of 1
The determination of solubilities of organometallic
complexes, as described here with 1 being used as an
example, yields reproducible data without any decom-
position of the complex being observed. It is applicable
to solids of low volatility and needs gram quantities of
the sample. The extraction method applied here (for
experimental details see Section 4.3) allows for the mea-
surement of solubilities of 1 up to concentrations of ap-
prox. 20 g lꢀ1. At higher concentrations, the expansion
tube of the back pressure regulator control may be
plugged by complex particles occur. Hence it was possi-
ble to measure the solubility of 1 within the technically
relevant range of 100 up to 237 bar, but not above
237 bar.
non-coordinating fluorinated phosphines (Fig. 4,
and ). Interestingly, the dependence of solubility on
¤
ꢁ
the CO2 density shows a similar characteristic curve that
exhibits a sharp bend for both complex 1 and the related
ligand P(4-CF3C6H4)3 [10]. The significance and the
cause of this bend are not yet clear and still under inves-
tigation. Compared to 1, the non-fluorinated complex
(Ph3P)2NiCl2 (m) [22] is poorly soluble in CO2 even at
a somewhat higher temperature, whereas the curve is
The appropriate flow rate for the extraction method
was established in experiments at 200 bar and 50 ꢀC
using CO2 flow rates of 1.25, 0.625, 0.313, and
0.125 ml minꢀ1. During the extraction, no decomposi-
tion of the complex was observed, as confirmed by IR
similar shaped to that of Ph P ( ) which again is much
3
n
better soluble in sc CO2 according to the other findings.
Note that with the method reported here solubility mea-
surement of the liquid phosphine L1 is experimentally
not feasible (see also Section 2.3).
PI
TI
E
F
M
D
G
CO2
Vent
A
B
C
Fig. 2. Extraction device consisting of: A, CO2 cylinder; B, syringe pump; C, oven; D, extraction cell; E, particle filter; F, back-pressure regulator;
G, sampling tube.

1470
N. Dahmen et al. / Journal of Organometallic Chemistry 690 (2005) 1467–1473
260
240
220
200
180
160
140
120
100
Flow rate = 1.250 ml/min,
Complex concentration =
1.12 mg/ml
Flow rate = 0.625 ml/min,
Complex concentration =
2.14 mg/ml
Flow rate = 0.3125 ml/min,
Complex concentration =
2.70 mg/ml
Flow rate = 0.125 ml/min,
Complex concentration =
2.84 mg/ml
0
20
40
60
80
100
120
140
CO2[ml] used for the extraction
Fig. 3. Extraction of 1 at 50 ꢀC in CO2 at varying flow rates.
Table 1
Solubility of 1 in CO2 (density 0.75 g cmꢀ3) at various temperatures
3. Conclusion
T (ꢀC)
p (bar)
Solubility (g lꢀ1
)
Solubility (mM)
The synthesis of [trans-Co2(CO)6{3,5-bis(CF3)-
C6H3P(i-C3H7)2}2] (1) and its solubility in supercritical
carbon dioxide is reported. The solubility of 1 is low
compared to non-coordinated phosphines, but high
compared to transition metal complexes bearing phos-
phines without fluorinated substituents. Solubility data
were determined between 40 and 70 ꢀC, but at higher
temperatures, however, a change of crystal structure or
melting of the compound could result in a remarkable
solubility change. Therefore, future measurements on
liquids, for example, on non-coordinated phosphines
or on solids which undergo melting induced by CO2
pressure, would require an alternative approach for
the determination of solubility data. The development
of an appropriate experimental setup for such measure-
ments is currently under way and will be reported in the
near future.
40
50
60
70
133.5
175.0
217.5
260.0
1.1 0.1
2.5 0.3
2.6 0.3
2.5 0.5
1.1 0.1
2.7 0.3
2.7 0.3
2.6 0.5
Table 2
Solubility of 1 at 50 ꢀC in CO2 at various pressures
p (bar) CO2 density (g cmꢀ3
)
Solubility (g lꢀ1
)
Solubility (mM)
100
125
150
175
200
213
225
237
0.38
0.61
0.70
0.75
0.78
0.80
0.81
0.82
<0.1
<0.1
0.7 0.1
2.1 0.2
2.5 0.3
2.7 0.1
7.0 0.7
11.4 0.4
15.8 0.8
0.8 0.1
2.2 0.2
2.7 0.3
2.8 0.1
7.4 0.7
12.1 0.4
16.7 0.9
1000
100
10
1
0.1
100
150
200
250
300
0.01
0.001
0.0001
0.00001
Pressure [bar]
Fig. 4. Solubility of (p-CF3C6H4)3P – ¤ (47 ꢀC) [10], Ph3P – n (47 ꢀC) [10], 1 –
(50 ꢀC), and (Ph3P)2NiCl2 – m (55 ꢀC) [22] in CO2 at varying
ꢁ
pressures.

N. Dahmen et al. / Journal of Organometallic Chemistry 690 (2005) 1467–1473
1471
4. Experimental
EI-MS, m/z (relative abundance): 330 (M+), 311
(C14H17F5P+), 288 (C11H11F6P+), 246 (C8H5F6P+), 227
(C8H5F5P+), 195 ðC8H4FþÞ 43 ðC3HþÞ. HR-EI-MS
4.1. Syntheses
5
7
(C14H17F6P), m/z (calculated): 330.0983 (330.0972).
All manipulations were performed under an atmo-
sphere of dry argon using standard Schlenk techniques.
3,5-Bis(CF3)C6H4-Br (ABCR) and (i-C3H7)2PCl (Al-
drich) were distilled prior to use. Co2(CO)8 (Aldrich)
was crystallized from pentane.
4.1.2. Co2(CO)6 [3,5-bis(CF3)C6H3P(i-C3H7)2] (1)
The solution of L1 (10.000 g, 30.27 mM) in 40 ml
benzene was added drop wise to the solution of
Co2(CO)8 (5.1778 g, 15.14 mM) in 60 ml benzene. The
reaction mixture was refluxed and stirred for 6 hours.
After cooling to 25 ꢀC, the solvent was removed in vac-
uum and the red coloured crude product (yield >95%)
was washed twice with n-pentane. 1 (MP 144 ꢀC, decom-
position) was crystallized from a 1:5 mixture of CH2Cl2/
pentane (68% yield).
The NMR spectra of 1 and L1 were recorded on a
1
Bruker Avance 250 spectrometer at 298 K. H NMR:
250.13 MHz, internal reference CDC13 (d = 7.27) rela-
tive to SiMe4 (d = 0 ppm); 13C{1H} NMR: 62.90 MHz,
internal reference CDC13 (d = 77.00) relative to SiMe4
(d = 0 ppm); 31P{1H} NMR: 101.25 MHz, internal ref-
erence 85% H3PO4 (d = 0 ppm). Chemical shifts and
coupling constants are given in ppm and Hz, respec-
tively. IR spectra were recorded on a Perkin–Elmer, Sys-
tem 2000 FT-IR, as a Nujol film (Nujol dried using K/
Na alloy) or in KBr (KBr dried at 500 ꢀC for several
hours). EI-MS (70 eV) spectra were obtained from
GC–MS analysis with a Hewlett–Packard 5890/5922
instrument. ESI-MS spectra were measured on a Hew-
lett–Packard Series 1100 MSD and HR-EI-MS on
Micromass GCT.
31P NMR: 90.3 (s). IR: 1973 w, 1952, vs, vb 1357 s,
1279 s, 1185 s, 1138 vs 1124 sh, 1096 w, 1052 vw, 905
w, 843 vw, 705 w, 683 w, 650 s, 536 w, 511 w, 500 w,
466 vw. ESI-MS, m/z (relative abundance): 984
(C34F12H35NaNO6P2Coþ; M+ + Na + NH3 ꢀ 2H), 968
2
(C34H33F12P2O6Co2Na+, M+ + Na ꢀ H), 903 (C31H26-
F12P2O6Coþ, M+ ꢀ C3H7), 369 (C14H19F6PNNa+;
L1 + Na ꢀ2H + NH3), 347 (C14H20F6PN+; L1 + NH3).
Melting or decomposition points were determined on
a Buchi B545 melting-point apparatus. Elemental analy-
¨
Table 3
Crystallographic data of 1 (standard deviations in parentheses)
ses were performed by Hermann Kolbe Micro Analyti-
cal Laboratory (Mulheim/Ruhr, Germany).
¨
Empirical formula
Formula weight
C34H34Co2F12O6P2
946.41
Crystal size (mm3)
Crystal system
0.45 · 0.45 · 0.1
Monoclinic
P21/c (no. 14)
4.1.1. (i-C3H7)2PC6H3(CF3)2, L1
Space group
Unit cell dimensions
A round flask with stirrer was charged with magne-
sium turnings (0.912 g, 37.50 mM) in 5 ml Et2O with
two drops of 1,2-dibromoethane. A solution of 3,5-
(CF3)2C6H3 Br (10.00 g, 34.13 mM) in 50 ml Et2O was
added drop wise to the magnesium turnings under stir-
ring. The reaction mixture was refluxed for two hours
and then filtrated. Within 1 h at 0 ꢀC, the Grignard solu-
tion was added to the solution of (i-C3H7)2PCl (5.21 g,
34.13 mM) in 40 ml diethyl ether. After warming up,
the mixture was hydrolyzed with degassed water. The
organic layer was separated, dried over MgSO4, and
the solvent was removed in vacuum. The dark brownish
liquid was distilled under vacuum at 60 ꢀC, yielding
10.0 g (90%) of L1 as a colorless liquid.
a
1019.79(17)
1293.2(2)
b
c
1610.1(3)
108.034(3)
2019.1(6) · 106
2
1.557
b
Volume (pm3)
Z
Dcalc (g cmꢀ3
Diffractometer
)
Siemens SMART 1000 CCD
diffractometer
Mo Ka, graphite monochromator
200
Wavelength
Temperature (K)
h Range (ꢀ)
Scan
2.06 6 h 6 28.31
x scan, Dx = 0.3ꢀ
ꢀ13 6 h 6 13, ꢀ16 6 k 6 17,
ꢀ21 6 l 6 21
Index ranges
3
3
Number of reflections measured
Independent reflections
Reflections observed
20697
1H NMR: 0.78 (dd, JH,H = 6.87, JH,P = 11.3 Hz,
4844
3974 (I > 2r)
290
3
3
6H, CH3), 0.96 (dd, JH,H = 7.12, JH, P = 15.25, 6H,
CH3), 2.04 (m, 2H, CH–P), 7.74 (s, 1H, Ph), 7.80 (d,
3JC-P = 5.5, 2H, Ph). 31P NMR: 13.38 (s). 13C NMR:
Number of parameters refined
Residual electron density (e pmꢀ3
Corrections
)
0.56 · 10-6
2
2
Lorentz and polarization,
exp. absorption correction
Direct methods
Full-matrix least-square on F2
18.30 (d, JC,P = 8.05, CH3), 19.26 (d, JC,P = 18.39,
1
CH3), 22.86 (d, JC,P = 13.79, CH–P), 122.57 (m, CH,
Structure solution
Structure refinement
Programs and weightings used
1
Ph), 123.67 (q, JC,F = 273.5, CF3), 131.40 (dq,
2JC,F = 36.6, JC, P = 6.90, C, Ph), 134.16 (d,
3
SHELX-97 [23], XPMA
ZORTEP [24]
,
2JC,P = 19.54, CH, Ph) 139.59 (d, JC,P = 27.58, C–P,
1
R indices
R1 = 0.0363 (I > 2r)
Ph). IR (Nujol): 1355 s, 1279 s, 1183 s, b, 1140 vb, vs,
1098 w, 1023 vw, 899 w, 843 vw, 708 w, 682 w.
Rw = 0.0992 (all data against F2)

1472
N. Dahmen et al. / Journal of Organometallic Chemistry 690 (2005) 1467–1473
20
18
16
14
12
10
8
6
4
2
0
100
105
110
115
120
125
Pressure [bar]
Fig. 5. Solubility of naphthalene in CO at 55 ꢀC. Own measurement – ¤, Chang/Morell [18] ꢁ, McHugh/Paulaitis – m, [25], Mitra et al. – [19],
n
2
Lamb et al. – q [26].
Elemental analysis: % found (calculated) C, 42.90
(43.13), H, 3.64 (3.62).
The separation method was examined in an experi-
ment with a known amount of triphenylphosphine being
completely extracted from the cell and collected in the
glass tube with an accuracy of 99.9%.
4.2. X-ray crystallography
In order to validate the method with regard to accu-
racy and reproducibility, solubility of naphthalene was
determined. As shown in Fig. 5, the values for the solu-
bility of naphthalene are in good agreement with other
published data [18,19,25,26].
Red crystals of 1 suitable for X-ray analysis were
grown from CH2Cl2/pentane solution. All diffraction
data were collected on a Siemens SMART 1000 CCD dif-
fractometer (Table 3). Crystallographic data of the
structure have been stored at the Cambridge Crystallo-
graphic Database Centre, Supplementary publication
Nos. CCDC 242209 (1). Copies of this information
may be obtained free of charge from The Director,
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK
(Fax: +44-1223-336033; e-mail: deposit@ccdc.cam.
ac.uk or www: http://www.ccdc.cam.ac.uk).
References
[1] B. Cornils, E. Kuntz, J. Organomet. Chem. 502 (1995) 177.
[2] E.D. Rogers, K.R. Seddon, Ionic Liquids as Green Solvents:
Progress and Prospects, American Chemical Society, Washington,
DC, 2003, Distributed by: Oxford University Press.
[3] P.G. Jessop, W. Leitner, in: P.G. Jessop, W. Leitner (Eds.),
Chemical Synthesis Using Supercritical Fluids, Wiley-VCH,
Weinheim, 1999, Chapter 4.7.1.
4.3. Solubility measurements
[4] R. van Eldik, F.-G. Kla¨rner, High Pressure Chemistry Synthetic,
Mechanistic, and Supercritical Applications, Wiley-VCH, New
York, 2002.
The solid (up to 2 g) is placed on a bed of glass beads
(0.25–0.5 mm) inside the extraction cell D (see Fig. 1,
Section 2.1) having a volume of 10 ml which is heated
to the temperature of measurement. To avoid the re-
moval of solid particles from the cell, a metal frit with
2 lm porosity was used.
The sc CO2 is compressed to the desired pressure
using the syringe pump B. The experiment proceeds by
pumping the sc CO2 at a constant flow rate through
the extraction cell. In the extraction cell, the solid dis-
solves in sc CO2 and is thereby removed from the cell.
The pressure is released through a back-pressure regula-
tor (F). As separator, specially designed glass tubes (G)
with frits were used. The separator is used to vent off the
gas and collect the particles precipitating while the sc
CO2 solution expands. Knowing the flow rate, the time
of sampling the solid, and the mass of the precipitated
solid, its solubility is calculated.
[5] P. Jessop, T. Ikariya, Y. Hsiao, T. Ikariya, R. Noyori, Chem.
Rev. 99 (1999) 475.
[6] W. Leitner, Acc. Chem. Res. 35 (2002) 746.
[7] N. Dahmen, S. Pitter, E. Dinjus, Green Chem. 5 (2003) 218.
[8] P. Webb, M. Sellin, T. Kunene, S. Williamson, A. Slawin, D.
Cole-Hamilton, J. Am. Chem. Soc. (2003) 15577.
[9] E. Beckman, J. Green, D. Hancu, Ind. Eng. Chem. Res. 41 (2002)
4466.
[10] K. Wagner, N. Dahmen, E. Dinjus, Chem. Eng. Data 45 (4)
(2000) 672.
[11] T. Davis, C. Erkey, Ind. Eng. Chem. Res. 39 (2000) 367.
[12] P.G. Jessop, W. Leitner, in: P.G. Jessop, W. Leitner (Eds.),
Chemical Synthesis Using Supercritical Fluids, Wiley-VCH,
Weinheim, 1999, Chapter 4.7.3.1 and references cited in.
[13] M.J. Chen, R.J. Klingler, J.W. Rathke, K.W. Kramarz, Organo-
metallics 23 (11) (2004) 2701.
[14] A.R. Mannig, J. Chem. Soc. (A) (1964) 634.
[15] R. Weber, U. Englert, B. Ganter, W. Keim, M. Mo¨thrath, Chem.
Commun. (2000) 1419.

N. Dahmen et al. / Journal of Organometallic Chemistry 690 (2005) 1467–1473
1473
[16] L. Brammer, J.C.M. Rivas, Ch.D. Spilling, J. Organomet. Chem.
609 (2000) 36.
[22] D. Paolo, C. Erkey, J. Chem. Eng. Data 43 (1998) 47.
[23] G.M. Sheldrick, SHELX-97, University Go¨ttingen, Germany,
1997.
[17] F.W.B. Einstein, R. Kirkland, Acta Cryst. B34 (1978) 1690.
[18] H. Chang, D. Morrell, J. Chem. Eng. Data 30 (1985) 74.
[19] S. Mitra, J.W. Chen, D. Viswanath, J. Chem. Eng. Data 33 (1988)
35.
[24] L. Zsolnai, XPMA
1997.
, ZORTEP, University Heidelberg, Germany,
[25] M. McHugh, M. Paulaitis, J. Chem. Eng. Data 25 (1980) 326.
[26] D. Lamb, T.M. Barbara, J. Jonas, J. Phys. Chem. 90 (1986)
4210.
[20] W. Cross, A. Akgerman, Ind. Eng. Chem. Res. 35 (1996) 1765.
[21] K. Liong, N. Foster, S. Ting, Ind. Eng. Chem. Res. 31 (1992) 400.