Synthesis and characterization of chromium and molybdenum(0) complexes bearing the (pentafluoroethyl)diphenylphosphine (pfepp) ligand

Article abstract of DOI:10.1016/j.jorganchem.2004.09.034

The synthesis and characterization of two Group VI pentacarbonyl complexes (pfepp)M(CO)₅ (M = Cr 1, Mo 2; pfepp = PPh₂C ₂F₅) are reported. Thermolysis of M(CO)₆ and pfepp in refluxing octane afforded 1 and 2 in moderate yields. These complexes were completely characterized by multinuclear NMR, IR and elemental analysis. X-ray structures for these complexes indicated they were isostructural, crystalizing in triclinic unit cells with four molecules per asymmetric unit. A comparison of the bond lengths in 1 and 2 to other (L)M(CO)₅ complexes showed a relationship between the M-C_ax bond length and the electronic influence of the phosphine ligand, and establishes the pfepp ligand as neither electron-rich nor electron-poor. A comparison of IR data with other (L)M(CO)₅ complexes also indicates the pfepp ligand is electronically neutral, with an electronic influence that approximates phosphites.

Full text of DOI:10.1016/j.jorganchem.2004.09.034

Journal of Organometallic Chemistry 690 (2005) 534–538
www.elsevier.com/locate/jorganchem
Note
Synthesis and characterization of chromium and
molybdenum(0) complexes bearing the
(pentafluoroethyl)diphenylphosphine (pfepp) ligand
a
Jason D. Palcic , Russell G. Baughman , Misha V. Golynskiy ,
b
b
a,
Sara B. Frawley a,c, R. Gregory Peters
*
a
Department of Chemistry, University of Memphis, Memphis, TN 38152, USA
b
Division of Science, Truman State University, Kirksville, MO 63501, USA
c
Division of Science, Keene State College, Keene, NH 03435, USA
Received 22 August 2004; accepted 14 September 2004
Abstract
The synthesis and characterization of two Group VI pentacarbonyl complexes (pfepp)M(CO)5 (M = Cr 1, Mo 2;
pfepp = PPh2C2F5) are reported. Thermolysis of M(CO)6 and pfepp in refluxing octane afforded 1 and 2 in moderate yields. These
complexes were completely characterized by multinuclear NMR, IR and elemental analysis. X-ray structures for these complexes
indicated they were isostructural, crystalizing in triclinic unit cells with four molecules per asymmetric unit. A comparison of the
bond lengths in 1 and 2 to other (L)M(CO)5 complexes showed a relationship between the M–Cax bond length and the electronic
influence of the phosphine ligand, and establishes the pfepp ligand as neither electron-rich nor electron-poor. A comparison of IR
data with other (L)M(CO)5 complexes also indicates the pfepp ligand is electronically neutral, with an electronic influence that
approximates phosphites.
ꢀ 2004 Published by Elsevier B.V.
Keywords: Chromium; Molybdenum; pfepp; Fluoroalkylphosphine; X-ray structure; IR
1. Introduction
pace. An a priori assessment of the stereoelectronic pro-
file of known phosphine ligands indicated several large
voids, which include large, electron-poor phosphines
and electronically moderate phosphines (neither elec-
tron-rich nor electron-poor), which confirms that devel-
opment of electronically neutral and/or electron poor
phosphines lags that of electron rich phosphines [2]. Gi-
ven the ubiquitous nature of phosphines in organome-
tallic chemistry and the incredible number of reactions
catalyzed by metal–phosphine complexes, these voids
are quite noteworthy in that they represent an opportu-
nity of phosphine ligand development and the potential
for discovery of novel reaction pathways.
Tertiary phosphines, PR3, are among the most uti-
lized group of ligands in transition metal chemistry,
due to the ease with which the steric and electronic prop-
erties of phosphine ligands are controlled [1]. Many of
the structure/reactivity relationships involving phos-
phines have focused on the introduction of traditional
hydrocarbon substituents to the phosphorus atom.
Phosphine ligands bearing hydrocarbon groups tend to
be viewed as electron rich, and the development of elec-
troneutral or electron-poor phosphines has not kept
With this consideration in mind, research in our
group has focused on the development of a new class
of phosphines of the type R2PRf, (R = hydrocarbon
*
Corresponding author. Tel.: +1 901 678 3293; fax: +1 901 678
3447.
E-mail address: rgpeters@memphis.edu (R.G. Peters).
0022-328X/$ - see front matter ꢀ 2004 Published by Elsevier B.V.
doi:10.1016/j.jorganchem.2004.09.034

J.D. Palcic et al. / Journal of Organometallic Chemistry 690 (2005) 534–538
535
substituent; Rf = C2F3 or C2F5). Phosphine ligands
bearing fluoroalkyl and fluorovinyl groups are readily
prepared via treatment of RfLi with an appropriate P–
Cl precursor [3–6]. We have already reported some of
our preliminary studies of Group VI hexacarbonyl sub-
stitution chemistry [7,8]. We have most recently reported
the synthesis and characterization of pentaflurorethyldi-
phenylphosphine (pfepp), prepared by treatment of
Ph2P–Cl with C2F5Li, and its coordination chemistry
with platinum [9,10].
ml of hexane and cooling to ꢀ78 ꢁC yielded 0.471 g
(65.2%) of Cr(CO)5(Ph2PC2F5) as a yellow solid. Anal.
Calc. for C19H10F5O5PCr: C, 45.99; H, 2.03. Found:
1
C, 45.86; H, 1.95%. H NMR (C6D6): d 7.55 (m, 2H),
6.92 (m, 3H). 19F NMR (CDCl3): d ꢀ76.5 (s, CF3);
2
ꢀ108.5 (d, JPF = 62 Hz, CF2). 31P NMR (C6D6): d
2
80.77 (t, JPF = 62 Hz, P–CF2). IR (nujol, cmꢀ1): 2072
(s), 1995 (m), 1954 (s), 1300 (m), 1218 (s), 1115 (m),
960 (m).
The present work is part of a continuing effort to pre-
pare and characterize electroneutral phosphine ligands.
In an effort to more broadly establish the electronic
influence of the pfepp ligand, we now wish to report
some of the substitution chemistry of Group VI hexa-
carbonyl complexes with pfepp. Treatment of M(CO)6
with one equivalent of pfepp in refluxing octane yielded
the desired pentacarbonyl complexes, M(CO)5(pfepp)
(M = Mo, Cr). IR and X-ray crystal structure data of
these complexes support the conclusion that the pfepp
ligand approximates the electronic influence of
phosphites.
2.3. Mo(CO)5(pfepp)
The procedure for this synthesis is analogous to the
Cr Complex except the reflux time was shortened to 5
h. Workup and isolation as before afforded 0.391 g
(48.2%) of Mo(CO)5(pfepp) as a white/light gray solid.
Anal. Calc. for C19H10F5O5PMo: C, 42.25; H, 1.86.
1
Found: C, 41.93; H, 2.10%. H NMR (C6D6): d 7.55
(m, 2H); 6.94 (m, 3H). 19F NMR (CDCl3): d ꢀ76.5 (s,
2
CF3); ꢀ109.5 (d, JPF = 65 Hz, CF2). 31P NMR
2
(C6D6): d 58.8 (t, JPF = 65 Hz, P–CF2). IR (nujol,
cmꢀ1): 2080 (s), 2000 (m), 1960 (s), 1300 (m), 1208 (s),
1115 (m), 960 (m).
2. Experimental
2.4. X-ray diffraction studies
2.1. General considerations
The crystallographic data for compounds 1 and 2 are
summarized in Table 2. Crystals of compound 1 were
isolated from an overnight room temperature evapora-
tion of a 1:1 mixture of diethyl ether and pet ether. Crys-
tals of 2 were obtained from a slow evaporation of a
petroleum ether solution. Single crystal X-ray data for
1 and 2 were collected on a Bruker P4 diffractometer
All manipulations were conducted under an inert
atmosphere using glove box, high-vacuum and/or Sch-
lenck techniques. Water and oxygen free solvents were
prepared from sodium/benzophenone and vacuum dis-
tilled prior to use. Metal hexacarbonyl starting materials
were obtained from Aldrich and used without further
˚
equipped with a molybdenum tube (k = 0.71073 A)
1
purification. H, 19F and 31P NMR spectra were meas-
and a graphite monochromator. Empirical absorption
corrections based on face indexing and integration were
applied; the structures were solved by direct methods
and refined by full matrix least squares techniques on
F2 using structure solution programs from the BRU-
ured using a JEOL 270 MHz spectrometer operating
at 270.17, 254.21 and 109.37 MHz, respectively. 31P
and 19F NMR spectra were externally referenced to
H3PO4 and CFCl3, respectively, with downfield shifts ta-
ken to be positive. Infrared spectra were obtained on a
Perkin–Elmer FTIR instrument as Nujol mulls. Elemen-
tal analyses were obtained from Desert Analytics.
PPh(C2F5)2 was prepared as described previously [9].
KER/SHELX 97 system. All nonhydrogen atoms were re-
fined anisotropically, while hydrogen atoms were placed
in calculated positions and refined with fixed isotropic
thermal parameters. Crystallographic data for the struc-
tural analysis of 1 and 2 have been deposited with the
Cambridge Crystallographic Data Centre, CCDC No.
247993 for 1 and CCDC No. 247994 for 2.
2.2. Cr(CO)5(pfepp)
Cr(CO)6 (0.320 g, 1.45 mmol) and 15 ml of octane
were combined in a 25 ml flask fitted with a reflux con-
denser and attached to a nitrogen manifold. To this
solution was added 0.459 g (1.51 mmol) of pfepp and
the solution was refluxed under nitrogen for 18 h. The
solution was transferred to a filtration assembly for
workup. The octane was removed under vacuum, the
residue was slurried in diethyl ether and filtered, remov-
ing traces of black residue assumed to be Cr(0) or Cr
oxides. Removal of ether followed by the addition of 2
3. Results and discussion
Thermolysis of M(CO)6 with one equivalent of pfepp
in refluxing octane yielded the monosubstituted penta-
carbonyl complexes M(CO)5(pfepp) (M = Cr 1 and
Mo 2) as seen in Eq. (1). Cr(CO)5(pfepp) was isolated
in moderate yield (65%) as a yellow solid after an ether
extraction from decomposed starting material followed

536
J.D. Palcic et al. / Journal of Organometallic Chemistry 690 (2005) 534–538
2
3
Table 1
Að12Þ IR (cmꢀ1) frequency data for Mo(CO)5 (L) complexes
59 ppm and a similar JPF of 65 Hz and no JPF cou-
pling. 19F data were relatively insensitive to the metal
center, as both complexes showed peaks typical for
CF3 and CF2 groups at ꢀ76 and ꢀ108 ppm,
respectively.
Ligand (L)
Mo(CO)5 (L)
PMe3
PPh3
2071 [13]
2073 [13]
2078 [15]
2078 [15]
2080
PPh2(CF@CF2)
P(OEt3)3
Ph2PC2F5
P(OPh)3
PBr3
A comparison of the IR CO stretching modes of these
compounds can be used as a qualitative indicator of the
relative electronic influence of these phosphines with the
metal center. Compounds of the type (R3P)M(CO)5 pos-
sess pseudo C4v symmetry and three IR active bands
attributable to CO stretches. The Að12Þ CO stretching
mode is an easily identified band that varies in a consist-
ent fashion, and depends on the relative donor ability of
the phosphine ligand. Table 1 shows a comparison of IR
stretching frequencies for a variety of (L)Mo(CO)5 com-
plexes and clearly establishes the electronic influence of
the pfepp ligand approximates phosphites and the per-
fluorovinyl analog PPh2(CF@CF2).
X-ray quality crystals of both 1 and 2 were grown
from the slow evaporation of a pentane or an ether/
pet ether solution. The complexes are isomorphous, pos-
sessing triclinic crystal symmetry with four molecules
per asymmetric unit (Z = 8) and an ORTEP drawing
of 1 is shown in Fig. 1. The presence of 4 molecules
per asymmetric unit was verified for 1 based on determi-
nation of identical lattice constants for three different
crystals and the variations in torsional angles of the four
conformers.
2083 [15]
2093 [15]
2095 [15]
PCl3
by isolation from hexane at ꢀ78 ꢁC. Similar workup of
the molybdenum analog yielded Mo(CO)5(pfepp) as a
white/tan solid, again in moderate yield (49%).
CO
CO
OC
OC
CO
CO
OC
OC
CO
∆ / R2P(Rf)
Mo
CO
Mo
CO
PR2Rf
ð1Þ
The presence of several spin active nuclei allowed
these complexes to be readily characterized by multinu-
clear NMR. 31P NMR for 1 showed a single resonance
at 82.2 ppm, a significant downfield shift from the free
2
ligand peak (ꢀ1.4 ppm) [9]. The JPF of 62 Hz for 1
was similar to the free ligand peak (58 Hz), though the
3JPF for 1 dropped to ꢁ0 Hz (3JPF for pfepp = 16.5
Hz). 31P NMR for 2 also showed a single resonance at
Bond length data from (L)M(CO)5 complexes is often
used as an indirect measure of the electronic character-
Table 2
Crystal data and structure refinement data for (CO)5Cr[PPh2(C2F5)] and (CO)5Mo[PPh2(C2F5)]
(CO)5Cr[PPh2(C2F5)]
(CO)5Mo[Ph2P(C2F5)]
Empirical formula
Formula weight
Crystal system
C19H10CrF5O5 P
496.24
Triclinic
C19H10F5O5MoP
540.18
Triclinic
ꢀ
ꢀ
Space group
Unit cell dimensions
P1
P1
˚
a (A)
12.0861(6)
12.2000(6)
28.382(2)
12.1674(9)
12.2604(12)
28.918(3)
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
92.676(4)
93.598(4)
92.837(5)
93.588(6)
90.493(4)
4171.9(4)
90.527(5)
4299.9(7)
3
˚
Volume (A )
Z
8
1.580
8
1.669
Dcalc (Mg/m3)
Absorption coefficient (mmꢀ1
F(000)
Crystal size (mm3)
)
0.697
1984
0.754
2128
0.58 · 0.46 · 0.40
1.79–25.00
16,882
0.50 · 0.40 · 0.23
1.77–25.03
17,430
Theta range for data collection (ꢁ)
Reflections collected
Independent reflections (Rint
)
14,614 (0.0224)
0.780 and 0.702
293(2)
15,069 (0.0236)
0.861 and 0.749
293(2)
Maximum and minimum transmission
Temperature (K)
Data/restraints/parameters
Final R indices [I > 2r(I)]
R indices (all data)
14,614/0/1117
R1 = 0.0392, wR2 = 0.0986
R1 = 0.0607, wR2 = 0.1087
15,069/0/1117
R1 = 0.0370, wR2 = 0.0.0957
R1 = 0.0532, wR2 = 0.1185

J.D. Palcic et al. / Journal of Organometallic Chemistry 690 (2005) 534–538
537
Table 3
M–Cax bond distance data (A) for (L)M(CO)5 complexes
˚
Ligand (L)
(L)Cr(CO)5
(L)Mo(CO)5
PMe3
PPh3
1.850(2) [11]
1.845(4) [16,17]
N/A
1.984(4) [11]
1.995(3) [14]
1.996(4) [7]
N/A
PPh2(CF@CF2)
P(OPh)3
1.861(4) [16]
1.870(4)
Ph2PC2F5
PCl3
2.012(5)
2.035(2) [12]
2.045
1.900(4) [11]
1.886
PF3 (calcÕd) [12]
the electronic impact of the bound phosphine ligand,
the M–Cax distance varies in a systematic fashion and
is readily related to the electronic character of the lig-
and. The M–Cax elongates as the phosphine ligand be-
comes more electron poor. This observation is readily
rationalized; by reducing the electron density at the me-
tal center the amount of electron density available for
overlap into the p* orbital of the CO ligand trans to
the phosphine is also reduced. The reduced overlap
weakens the M–C bond, which slightly elongates the
bond. Table 3 shows a list of M–Cax bond distances
for a number of (L)Mo(CO)5 and (L)Cr(CO)5 com-
plexes. In both series, the longest bond belongs to the
Fig. 1. ORTEP drawing of (pfepp)Cr(CO)5. Hydrogen atoms have
˚
been omitted for clarity. Selected bond length (A) and angle (ꢁ) data
for 1 and 2: (pfepp)Cr(CO)5 1, Cr–P(1) 2.3663(9), Cr–C(1)axial 1.870(4),
Cr–C(eq)av 1.903(4), C(1)–Cr–P(1) 176.14(11). (pfepp)Mo(CO)5 2.
Mo–P(1) 2.5106(11), Mo–C(1)axial 2.012(5), Mo–C(eq)av 2.049(5),
C(1)–Mo–P(1) 174.83(14).
˚
PCl3 derivative (1.900(4) A for (PCl3)Cr(CO)5 and
˚
2.035(2) A for (PCl3)Mo(CO)5) [11,12] and the shortest
istics of a phosphine ligand. The Cr–P bond distance in
˚
1 of is 2.3663(9) A and the Mo–P bond length of 2 is
˚
2.5016(11) A. The Cr–P bond length of (pfepp)Cr(CO)5
is essentially identical to the Cr–P bond length of
˚
2.3664(5) A in (PMe3)Cr(CO)5 [11] and is only slightly
˚
shorter than the Cr–P bond of 2.422(1) A in the PPh3
bond occurs in the PMe3 derivative (1.850(2) and
1.984(4) for the Cr and Mo complexes, respectively)
[11]. The M–Cax bond length data for 1 and 2
˚
˚
(1.870(4) A for 1 and 2.012(5) A for 2) clearly show
the electronic influence of the pfepp ligand approximates
phosphites and PPh2(CF@CF2).
analog [12,13]. Similar comparisons between the Mo–P
bond distances in 2 and other (L)Mo(CO)5 complexes
indicates there is little correlation between the electronic
influence of a phosphine ligand and the metal–phospho-
The metal centers of 1 and 2 sit in the center of a
slightly distorted octahedron. The four equatorial car-
bonyls are planar but the metal center lies out of this
plane, puckered slightly towards the phosphorus ligand.
Consequently, all four Cax–M–Ceq bonds are slightly
less then 90ꢁ. The P–M–Cax bond angle of 176.14ꢁ is
slightly deflected from the idealized 180ꢁ. This deflection
of the M–P bond results in P–M–C3 (and C4) bond an-
gles that are slightly less than 90ꢁ and P–M–C2 (and C5)
bond angles that are slightly greater than 90ꢁ.
˚
rus bond. The Mo–P bond length of 2 is 2.5016(11) A is
again almost identical to the Mo–P bond distance of
˚
several (L)Mo(CO)5 complexes [L = PMe3 (2.5082 A)
˚
[11], P(CH2CH2CN)3 (2.506 A) [14], PPh2(CF@CF2)
˚
(2.5168 A)] [7] bearing phosphines with varying elec-
tronic characteristics.
The equatorial metal carbonyl bonds are also rela-
tively insensitive to the electronic nature of the phos-
phine ligand. The average M–Ceq for (PF3)Mo(CO)5 is
4. Conclusions
˚
2.045 A [12] and the average bond length for the
˚
(PMe3)Mo(CO)5 analog is 2.036 A [11]. Similar compar-
Two new Group VI pentacarbonyl complexes
(pfepp)M(CO)5 (M = Cr 1, Mo 2; pfepp = PPh2C2F5)
were prepared and characterized by IR, NMR, EA
and X-ray crystallography. A comparison of IR stretch-
ing frequencies and M–Cax bond length data with other
(L)M(CO)5 complexes indicates the pfepp ligand is an
electroneutral phosphine, with an electronic influence
that approximates phosphites. The pfepp ligand occu-
pies one of the two voids in the stereoelectronic profile
of typical phosphine ligands.
isons for the chromium complexes also show no
dependence of the M–Ceq bond length with the phos-
phine ligand. The average M–Ceq for (PF3)Cr(CO)5 is
˚
1.886 A [12] and the bond length for the (PMe3)Mo-
˚
(CO)5 analog is 1.850(2) A [11]. The average M–Ceq
˚
bond length in the pfepp derivatives is 1.903 A for 1
˚
and 2.049 A for 2.
While the metrical data from M–P and M–Ceq bond
distances for (L)M(CO)5 complexes is independent of

538
J.D. Palcic et al. / Journal of Organometallic Chemistry 690 (2005) 534–538
[6] R.G. Peters, B.L. Bennett, R.C. Schnabel, D.M. Roddick, Inorg.
Chem. 36 (1997) 5962–5965.
Acknowledgements
[7] R.G. Peters, J.D. Palcic, R.G. Baughman, Acta Crystallogr. E59
(2003) m1198–m1200.
The authors acknowledge the support of this work by
two National Science Foundation Research Experience
for Undergraduates (REU) grants (CHE03-53885,
SBF; and CHE99-87775, MVG). This work was also
supported in part by a University of Memphis Faculty
Research Grant.
[8] R.G. Peters, M.V. Golynskiy, R.G. Baughman, Acta Crystallogr.
E58 (2002) m70–m71.
[9] J.D. Palcic, P.N. Kapoor, D.M. Roddick, R.G. Peters, J. Chem.
Soc., Dalton Trans. (2004) 1644–1647.
[10] J.D. Palcic, R.G. Baughman, R.G. Peters, J. Coord. Chem.
(submitted).
[11] M.S. Davies, M.J. Aroney, I.E. Buys, T.W. Hambley, J.L.
Calvert, Inorg. Chem. 34 (1995) 330–336.
References
[12] G. Frenking, K. Wichmann, N. Froehlich, J. Grobe, W. Golla, D.L.
Van, B. Krebs, M. Laege, Organometallics 21 (2002) 2921–2930.
[13] M.S. Davies, G.W. Allen, M.J. Aroney, T.W. Hambley, R.K.
Pierens, J. Mol. Struct. 326 (1994) 81–91.
[1] C.A. Tolman, Chem. Rev. 77 (1977) 313–348.
[2] K.D. Cooney, T.R. Cundari, N.W. Hoffman, K.A. Pittard, M.D.
Temple, Y. Zhao, J. Am. Chem. Soc. 125 (2003) 4318–4324.
[3] K.K. Banger, A.K. Brisdon, A. Gupta, Chem. Commun. (Cam-
bridge) (1997) 139–140.
[14] F.A. Cotton, D.J. Darensbourg, W.H. Ilsley, Inorg. Chem. 20
(1981) 578–583.
[15] N.A. Barnes, A.K. Brisdon, M.J. Ellis, R.G. Pritchard, J. Fluorine
Chem. 112 (2001) 35–45.
[4] K.K. Banger, R.P. Banham, A.K. Brisdon, W.I. Cross, G.
Damant, S. Parsons, R.G. Pritchard, A. Sousa-Pedrares, J. Chem.
Soc., Dalton Trans. (1999) 427–434.
[16] H.J. Plastas, J.M. Stewart, S.O. Grim, Inorg. Chem. 12 (1973)
265–272.
[5] M.F. Ernst, D.M. Roddick, Inorg. Chem. 28 (1989) 1624–
1627.
[17] H.J. Plastas, J.M. Stewart, S.O. Grim, J. Am. Chem. Soc. 91
(1969) 4326–4327.