Electrochemistry and [60]fullerene displacement reactions of (dihapto-[60]fullerene) pentacarbonyl metal(0) (M = Cr, Mo, W)

Article abstract of DOI:10.1039/b618260g

Cyclic voltammetry (CV) measurements on (η²-C ₆₀)M(CO)₅ complexes (M = Cr, Mo, W) in dichloromethane show three [60]fullerene-centered and reversible reduction/oxidation waves. The E_1/2 values of these waves are shifted to positive values relative to the corresponding values of the uncoordinated [60]fullerene in the same solvent. A Jahn-Teller type distortion of the spherical surface of [60]fullerene promoted by [60]fullerene-metal π-backbonding may explain the observed positive shifts. Lewis bases (L = piperidine and triphenyl phosphine) displace [60]fullerene from (η²-C₆₀)M(CO)₅ complexes. Analysis of the activation parameters for the metal-[60]fullerene dissociation, the metal-[60]fullerene bond enthalpies (from DFT computations), and metal-solvent (benzene) bond enthalpies (from DFT computations) suggests appreciable solvent contribution to the transition state leading to formation of the intermediate species solvent-M(CO)₅. Appreciable transition state stabilization due to solvation of the intermediate species is inferred for M = Mo and W. For M = Cr, stabilization of the intermediate species due to solvation is not accompanied by the corresponding transition state stabilization. The Royal Society of Chemistry 2007.

Full text of DOI:10.1039/b618260g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Electrochemistry and [60]fullerene displacement reactions of
(dihapto-[60]fullerene) pentacarbonyl metal(0) (M = Cr, Mo, W)†
Elvin Igartu´a-Nieves, Yessenia Ocasio-Delgado, Mar´ıa D. L. A. Torres-Castillo,‡ Omar Rivera-Betancourt,‡
Jose´ A. Rivera-Paga´n,§ Diana Rodriguez,‡ Gustavo E. Lo´pez and Jose´ E. Corte´s-Figueroa*
Received 14th December 2006, Accepted 31st January 2007
First published as an Advance Article on the web 9th February 2007
DOI: 10.1039/b618260g
2
Cyclic voltammetry (CV) measurements on (g -C60)M(CO)5 complexes (M = Cr, Mo, W) in
dichloromethane show three [60]fullerene-centered and reversible reduction/oxidation waves. The E1/2
values of these waves are shifted to positive values relative to the corresponding values of the
uncoordinated [60]fullerene in the same solvent. A Jahn–Teller type distortion of the spherical surface
of [60]fullerene promoted by [60]fullerene–metal p-backbonding may explain the observed positive
2
shifts. Lewis bases (L = piperidine and triphenyl phosphine) displace [60]fullerene from (g -C60)M(CO)5
complexes. Analysis of the activation parameters for the metal–[60]fullerene dissociation, the
metal–[60]fullerene bond enthalpies (from DFT computations), and metal–solvent (benzene) bond
enthalpies (from DFT computations) suggests appreciable solvent contribution to the transition state
leading to formation of the intermediate species solvent–M(CO)5. Appreciable transition state
stabilization due to solvation of the intermediate species is inferred for M = Mo and W. For M = Cr,
stabilization of the intermediate species due to solvation is not accompanied by the corresponding
transition state stabilization.
In addition, the formation of [60]fullerene–metal complexes may
Introduction
be favored by the ability of [60]fullerene to participate in p-back
bonding.1–4 The [60]fullerene–metal bond dissociation energies
Buckminsterfullerene ([60]fullerene or C60) reacts with metal
carbonyl complexes to form [60]fullerene–metal carbonyl
complexes.1–19 The dihapto (g ) mode of [60]fullerene–transition
2
have been estimated in (g -C60)M(PH3)2 (M = Ni, Pd, Pt) com-
2
plexes by density functional calculations.32–34 The preparation and
electrochemical studies of some [60]fullerene-substituted metal
carbonyl complexes of group 6 have been reported.1–11 However,
to our knowledge, there are not reports on the plausible relation
between their reactivities toward [60]fullerene–ligand exchange
reactions and their electrochemical profiles.
metal coordination resembles an olefin–metal bond.1–16 Although
all carbons in [60]fullerene are equivalent, the molecule possesses
two distinct types of C–C bonds (ring : ring junctions): the C–C
bonds between six-membered rings (6 : 6 junction) and the C–
C bonds between six-membered and five-membered rings (6 : 5
junctions).20–22 The C–C bonds at 6 : 6 junctions are shorter than
the C–C bonds at 6 : 5 junctions and behave like olefinic units.
The existence of triply-degenerate low lying LUMOs capable of
accepting six electrons may account for the high electron affinity of
[60]fullerene.23–29 Six reversible reduction waves at potentials rang-
ing from −0.50 V to −3.26 V (vs. ferrocene/ferrocenium, Fc/Fc+)
have been reported for [60]fullerene.23, 24 Despite the potential of
[60]fullerene to bind transition metals in a variety of hapticities,
In this work we are reporting the results on the preparation,
electrochemical, kinetics and mechanistic studies of the ligand ex-
2
change reactions on (g -C60)M(CO)5 (M = Cr, Mo, W) complexes.
Experimental
General
the dihapto (g ) mode of coordination is preferred.1–15 Higher
2
Thin layer chromatography was performed using silica gel (25 lm,
˚
pore size 60 A, Aldrich) and UV light at 254 or 366 nm was
hapticities in aryl-substituted fullerides have been reported but are
scarce.17–19 The orbitals on the [60]fullerene spherical and rigid sur-
face are spatially oriented in such a way that hapticities higher than
two are not favored.30 The C–C double bonds at the 6 : 6 junctions
on the curved surface of [60]fullerene are strained and its release
upon forming a [60]fullerene–metal bond may explain the reactiv-
ity of [60]fullerene toward [60]fullerene–metal bond formation.31
used to visualize the components. Column chromatography was
performed using a 15 cm long (1 cm diameter) column packed
˚
with 62 grade, 60–2000 mesh, 150 A silica gel purchased from
Aldrich. Infrared spectra were obtained on a Bruker Vector 22
Fourier transform infrared spectrophotometer and UV-visible
spectra on a Perkin Elmer Lambda 25 UV/vis spectrophotometer.
All reactions were carried out under an inert nitrogen atmosphere.
A Julabo F 12-EC model heating and refrigerating circulator
and a K/J Fluke digital thermometer equipped with a bead
thermocouple were used as temperature control devices.
Electrochemical studies were performed at ambient temperature
and low pressure using a BAS CV-50W voltammetric work
station. A high vacuum line was used to transfer and mix
Organometallic Chemistry Research Laboratory, Department of Chemistry,
University of Puerto Rico, P. O. Box 9019, Mayagu¨ez, Puerto Rico, 00681-
9019
† The HTML version of this article has been enhanced with colour images.
‡ Participant of the University of Puerto Rico Undergraduate Research
Program.
§ ACS-Petroleum Research Fund Summer Research Fellow
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1293–1299 | 1293
©

View Article Online
reagents. Dichloromethane (Aldrich), used as solvent, was dried
over phosphorus pentoxide, fractionally distilled under nitrogen,
transferred and stored over dried silica under vacuum. The
silica was dried by heating with a bunsen burner under high
vacuum. Tetrabutylammonium hexafluorophosphate (TBPF6)
(Aldrich), used as supporting electrolyte, was recrystallized from
an ethanol/H2O (95 : 5) solution and dried under vacuum prior
to use. Decamethylferrocene (Fc*)/decamethylferrocenium (Fc*+)
couple was used as internal standard in all measurements.
A three-electrode configuration was used consisting of a glassy
carbon working electrode (3 mm in diameter), a platinum wire
(Pt-wire) counter electrode, and a non-aqueous silver pseudo-
reference electrode. The non-aqueous silver pseudo-reference
electrode consists of a silver wire inside a 4 mm (O. D.) Pyrex tubing
which contains a small amount of a 0.1 M TBPF6 solution in
dichloromethane and separated from the bulk solution by a Vycor
glass frit located at the tip of the tube. This electrode configuration
prevents peak potential drift due to small changes in the solution
ionic strength during the electrochemical measurements. The
working electrode was polished before use with a 0.25 lm diamond
polishing compound (Metadi II) purchased from Buehler and a
microcloth (BAS) and sonicated for a few seconds in deionized
water to remove adsorbed polishing materials. The Pt-wire was
cleaned by exposing it to a flame for approximately 30 s and the
silver wire was rinsed with acetone and deionized water to remove
impurities.
sample voltammogram versus the internal reference, the solution
was then transferred over decamethylferrocene and the CV was
scanned in the same voltage range. When decamethylferrocene was
not added at the beginning of the experiment, the cell was opened
under argon and a small amount of the reference was placed in
one of the cell’s arm. The vacuum was then re-established slowly.
Preparation and purification of materials
Benzene (Fisher) was dried over sodium metal for 24 h and frac-
2
tionally distilled under nitrogen. The complexes (g -C60)M(CO)5
were prepared photochemically from M(CO)6 (Aldrich) and
[60]fullerene (Aldrich) using a medium pressure mercury lamp
following a published procedure.4,5 Each complex was identified
by its infrared spectrum in the carbonyl stretching region (mco). The
mco of the complexes dissolved in n-hexane showed the expected mco
2
bands for a C4v local symmetry: mCO for (g -C60)W(CO)5; 2085 cm−1,
2
1996 cm−1, and 1972 cm−1; mCO for (g -C60)Mo(CO)5; 2089 cm−1,
2
1984 cm−1, and 1970 cm−1; mCO for (g -C60)Cr(CO)5; 2079 cm−1,
1987 cm−1, and 1969 cm−1.
Kinetics experiments
2
The rates of (g -C60)M(CO)5 disappearance were monitored by
observing the decrease of the absorbance values at 407 nm, 430 nm,
and 425 nm for M = Cr, Mo, and W, respectively. The reactions
were studied under flooding conditions where
A custom-made electrochemical cell was used. The cell is
equipped with two joints and two special arm adapters which
allowed sequential addition of sample, of the internal reference
and other compounds as necessary. One of the joints is used as a
vacuum line attachment and the other joint is used for electrodes
connections. The sample of the compound to be studied was
placed in one of the arm adapters (enough to obtain a solution of
approximately 0.5 mM to 1.0 mM in 3 mL of dichloromethane).
A minute amount of the internal reference, decamethylferrocene,
can be placed at the beginning of the experiment in the other
arm adapter. If the internal reference is placed at the beginning
of the experiment, a trace will sublimate into the solution during
the solvent-transfer process and a small wave due to Fc*/Fc+*
will show in the background. The supporting electrolyte (ca.
0.12 g of TBPF6) and a small spinning bar was placed in the
cell and connected to the vacuum line. The cell was opened
to the vacuum line (10−5 to 10−6 mmHg) and the supporting
electrolyte was heated with a heat gun for approximately five
seconds followed by cooling at room temperature. The heating
and cooling process was repeated five times to ensure that the
supporting electrolyte was completely dry. Approximately 3 mL
of dichloromethane were transferred to the cell directly through
the vacuum line by cooling the cell with liquid nitrogen. Once the
solvent transfer was accomplished, the whole cell ensemble was
disconnected from the vacuum line and allowed to reach room
temperature. The background voltammogram, dichloromethane
and supporting electrolyte was recorded before the sample electro-
chemical measurement at a scan rate of 100 mV s−1 between 0 mV
and −2000 mV. When determination of the cyclic voltammogram
involved multiple scans, the solution was stirred between each
scan using the spinning bar placed in the cell. Then, the supporting
electrolyte solution was transferred over the sample and a complete
scan was carried out from +1000 mV to −2000 mV. To record the
the concentrations of the ligand, L, were at least two hun-
dred times greater than the concentration of (g -C60)M(CO)5 (ca.
2
10−5 M) and under conditions where 0 ≤ [C60]/[L] ≈1 (for M =
W; L = PPh3, P(OEt)).4 The pseudo-first order rate constant
values were determined using a non-linear curve fitting of the
absorbance versus time plots. Error limits are given in parentheses
as the uncertainties of the last digit. Under conditions where
[C60]/[L] ≈0, the rate constant values were independent of the
ligand concentration, [L].
Computational method
The computational model used in this study is based on ab initio
techniques within the density functional approximation (DFT).
Specifically, the Becke three parameter hybrid functional termed
B3LYP was used with the LanL2DZ Gaussian basis set. For
all systems studied a full geometry optimization was performed.
Various initial geometries in the optimization procedure were used
to guarantee the determination of the lowest energy equilibrium
structure. All calculations were performed using Gaussian 0335
installed on an SGI 2000 computer. The binding energy of each
complex, DEX–Metal(CO)5, was computed using the supermolecule
approach, namely
DEX–Metal(CO) = EX–Metal(CO) − EMetal(CO) − EX
5
5
5
where DEK represent the energy of the K species (K = X–
metal(CO)5). In this case X will be either the solvent molecule
(benzene) or fullerene, and the metal will be Cr, Mo, and W. Pre-
liminary calculations were performed to determine the magnitude
of the basis set superposition error (BSSE) for these systems using
the counterpoise correction. In all cases the magnitude of the
BSSE was less than 1 kJ mol−1 which compared to the values of
the binding energies obtained was negligible. The bond enthalpy
1294 | Dalton Trans., 2007, 1293–1299
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
was computed from the binding energy by correcting for zero point
energy, which in all cases was relatively small when compared to
the binding energy.
Results and discussion
Electrochemical experiments
2
Each (g -C60)M(CO)5 complex and C60 exhibits three reversible
one-electron reduction waves (Fig. 1–4). The half wave potential
values for each complex and for uncoordinated [60]fullerene
are presented in Table 1. The observed reductions are reversible
suggesting that they are [60]fullerene-centered. Except for the
2
second and third reduction waves of (g -C60)Mo(CO)5, they are
all shifted to positive potentials relative to the corresponding
potentials of the uncoordinated [60]fullerene. This observation is
at variance with the observed negative shift potential (cathodic
shift) for other [60]fullerene–M (M = Mo, W) complexes
containing non-CO co-coordinated ligands.8–11 For example,
observation of cathodic shifts of approximately 0.2 V for
Fig. 1 Cyclic voltammetric responses recorded at a glassy carbon working
electrode on dichloromethane solution containing 0.1 M TBPF6 and: (a)
C60 (saturated solution) and traces of decamethylferrocene (Fc*). Scan
rate 100 mV s−1. E1/2 for Fc*/Fc+* is 339 mV. The E1/2 for C60 are −998,
−1391 and −1860 mV vs. Fc*/Fc+*. T = 20 ◦C.
2
2
2
M(g -C60)(CO)2(phen)(dbm),8,9 M(g -C60)(CO)3(dppe),10 M(g -
2
C60)(CO)3(dppb),10 and M(g -C60)(CO)3(dppf)11 complexes
(phen = 1, 10-phenanthroline, dbm = dibutyl maleate, dppe =
bis-(1,2-diphenylphosphino)ethane, dppb = bis-(1,2-diphenyl-
phosphino)benzene, dppf = 1, 1ꢀ-(diphenylphosphino)ferrocene)
suggests that the extent of the shift is independent of the
number and nature of co-coordinated ligands. Positive shifts
2
2
are reported for the iron complexes [(g -C60)(S)2Fe2(CO)6], [(g -
2
C60)(S)2Fe2(CO)6]2, [(g -C60)(S)2Fe2(CO)6]316 It has been suggested
that the inductive effect exerted by the metal on [60]fullerene
=
and the disruption of C C double bond conjugation caused by
[60]fullerene p-back electron acceptance are opposed effects that
cancel each other to some extent. The inductive effect can either
disfavor or favor reduction of [60]fullerene, whereas disruption
=
of C C conjugation by [60]fullerene p-back electron acceptance
disfavors reduction. In the present study the first one-electron
2
reduction in each (g -C60)M(CO)5 complex is shifted to positive
values with respect to uncoordinated [60]fullerene. In the case
2
of (g -C60)Mo(CO)5, a cathodic shift is observed in the second
reduction wave (DE21/2 = −20 mV), while the cathodic shift of the
third reduction band is small (DE31/2 = −2 mV). The shifts on the
Fig. 2 Cyclic voltammetric responses recorded at a glassy carbon working
electrode on dichloromethane solution containing 0.1 M TBPF6 and: (a)
2
2
reductions of (g -C60)M(CO)5 complexes are positive despite the
(g -C60)Cr(CO)5 (saturated solution) and traces of decamethylferrocene
(Fc*). Scan rate 100 mV s−1. The E1/2 for Fc*/Fc+* is 385 mV. The E1/2
fact that [60]fullerene is a good p-acceptor. This positive potential
shift may be attributable to metal coordination. Donation of
electron density into the triply-degenerate [60]fullerene LUMOs
should lower the energy of at least one of the triply-degenerate
LUMOs so that reduction of the [60]fullerene moiety in
[60]fullerene–M complexes occurs at a lower energy relative to
uncoordinated [60]fullerene.36 No direct correlation was found
between the DE1/2 values with the M–L (L = C60, benzene) bond
2
for (g -C60)Cr(CO)5 are −996, −1380 and −1839 mV vs. Fc*/Fc+*. T =
20 ◦C.
enthalpies nor with the carbonyl stretching wavenumber of the
2
carbonyl trans to C60 in (g -C60)M(CO)5 complexes.
Displacement of C60 from (g2-C60)M(CO)5
=
The Lewis bases ( L) triphenyl phosphine and piperidine displace
[60]fullerene from (g -C60)M(CO)5 to produce (g -L)M(CO)5 (eqn
2
2
1
Table 1 En
values for C60 and (g -C60)M(CO)5 complexes in
dichloromethane
1/2
(1)).
E11/2, red/mV
E21/2, red/mV
E31/2, red/mV
2
1
(g -C60)M(CO)5 + L → (g -L)M(CO)5 + C60
(1)
2
The rates of (g -C60)M(CO)5 disappearance were monitored by
observing the decrease of the absorbance values at various
wavelengths, depending on the complex (see Experimental sec-
tion). The reactions were studied under flooding conditions
C60
−998
−996
−977
−982
−1391
−1380
−1411
−1374
−1860
−1839
−1862
−1844
2
(g -C60)Cr(CO)5
2
(g -C60)Mo(CO)5
2
(g -C60)W(CO)5
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1293–1299 | 1295
©

View Article Online
Table 2 Average rate constant values for [60]fullerene displacement
2
from (g -C60)M(CO)5 by L (L = pip and PPh3) in benzene at various
2
temperatures under flooding conditions such that [L] ꢁ[(g -C60)M(CO)5]
and [C60] ≈0
Temp./K 0.1 Metal
Ligand
Solvent
benzene
104 kobsd/s−1
293.2
298.2
303.2
298.2
303.2
308.2
317.3
327.3
337.2
317.3
327.3
337.3
Cr
Mo
W
piperidine
34(2)
61(3)
98(4)
37(2)
69(3)
121(6)
1.31(7)
2.7(1)
11(2)
1.01(5)
3.0(2)
11.9(6)
piperidine
piperidine
PPh3
benzene
benzene
benzene
W
where the concentrations of L were at least two hundred times
Fig. 3 Cyclic voltammetric responses recorded at a glassy carbon working
2
greater than the concentration of (g -C60)M(CO)5 (ca. 10−5 M)
electrode on dichloromethane solution containing 0.1 M TBPF6 and: (a)
and under conditions where 0 ≤ [C60]/[L] ≈1 (for M = W; L =
PPh3, P(OEt)).4 The pseudo-first order rate constant values were
determined using a non-linear curve fitting of the absorbance
versus time plots. Under conditions where [C60]/[L] ≈0, the rate
constant values were independent of the ligand concentration,
[L]. The rate constant values determined for various [L] and
temperatures under conditions that [C60]/[L] ≈0 are presented
in Table 2. The corresponding activation parameters, presented in
Table 3, were determined from Eyring plots.
2
(g -C60)Mo(CO)5 (saturated solution) and traces of decamethylferrocene
(Fc*). Scan rate 100 mV s−1. The E1/2 for Fc*/Fc+* is 385 mV. The E1/2
2
for (g -C60)Mo(CO)5 are −977, −1411 and −1862 mV vs. Fc*/Fc+*. T =
20 ◦C.
Interestingly for L = piperidine, the reactions were triphasic.
The absorbance values decreased with time in the first segment of
the triphasic plots. In the second and third segments of the plots,
the absorbance values increased with time, Fig. 5. The decreasing
segment of the plots was ascribed to displacement of [60]fullerene
2
1
from (g -C60)M(CO)5 producing (g -pip)M(CO)5 (eqn (2)),
while the second and third segments (increasing segments) were
assigned to step-wise additions of piperidine to uncoordinated
[60]fullerene (eqn (3)).
Fig. 4 Cyclic voltammetric responses recorded at a glassy carbon working
electrode on dichloromethane solution containing 0.1 M TBPF6 and: (a)
2
(g -C60)W(CO)5 (saturated solution) and traces of decamethylferrocene
(Fc*). Scan rate 100 mV s−1. The E1/2 for Fc*/Fc+* is 387 mV. The E1/2
2
for (g -C60)W(CO)5 are −1016, −1408 and −1882 mV vs. Fc*/Fc*+. T =
20 ◦C.
(3)
1296 | Dalton Trans., 2007, 1293–1299
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
2
Table 3 Activation parameters for [60]fullerene dissociation from (g -C60)M(CO)5 complexes in various solvents, (CO)5M–C60 and (CO)5M–benzene
bond enthalpies
(CO)5M–C60 bond
(CO)5 M–bz bond
enthalpy/kJ mol−1
Relative (M–bz/M–C60) bond strength
in [(CO)5M(bz)(C60)] transition state
M
Solvent
DH1‡/kJ mol−1
DS1‡/J K−1 mol−1
−35(8)
enthalpy/kJ mol−1 (DFT)
Cr
bz
bz
bz
bz
75(3)
78
57d
12/59 = 0.20
57(3)e
48(5)f
39.3(4)g
66d
Mo bz
88(1)
4(4)
101
131
52.0(2)h
76d
W
bza
bzb
bzc
CS2
107(6)
105(1)
92(13)
101.7(4)
88(3)
17(17)
16(4)
−26(26)
+20(1)
−30(9)
−9(9)
2.3/68 = 0.03
62(2)h
4
W
W
W
131
131
131
Clbz4
tol4
97(3)
bz = benzene, Clbz = chlorobenzene, tol = toluene.a This work for entering ligand L = PPh3. b Entering ligand L = PPh3, reference 4. c This work for
entering ligand L = piperidine. d This work from DFT. e From gas phase kinetics studies, ref. 45. f From condensed-phase kinetics studies, ref. 44. g From
condensed phase kinetics studies, ref. 43. h From condensed phase kinetics studies, ref. 46.
Fig. 5 Fig. 1. Plot of absorbance (407 nm) vs. time (s) for consecutive
Fig. 6 Plot of absorbance (407 nm) vs. time (s) for addition of piperidine to
C60 in chlorobenzene 328.2 K under flooding conditions where [piperidine]
ꢁ [C60] and [piperidine] = 1.4798 M. The continuous trace is for
absorbance values obtained from the equation that best describes the
2
displacement of C60 from (g -C60)Cr(CO)5 by piperidine in chlorobenzene
1
to form (g -pip)Cr(CO)5 and C60(C5H5N)3(H)3 at 328.2 K under flooding
2
conditions where [pip] ꢁ [(g -C60)Cr(CO)5]. Displacement of [60]fullerene
2
1
1 t
2 t
from (g -C60)Cr(CO)5 producing (g -pip)Cr(CO)5 was ascribed to the
decreasing segment of the plot. The first and second increasing segments of
the plot are assigned to step-wise additions of piperidine to uncoordinated
[60]fullerene.
relation between absorbance and time. At = ae−k − be−k + A∞ a =
0.03694, b = 0.02079, kꢀobsd = 3.11 × 10−3 s−1, kꢀꢀ
A∞ = 0.1420.
= 5.69 × 10−4 s−1 and
obsd
The activation parameters are consistent with an interchange-
2
dissociative displacement of [60]fullerene from (g -C60)M(CO)5,
where the solvent may assist the rupture of the [60]fullerene–metal
bond (scheme 1, path 1). Unassisted dissociation of [60]fullerene
(Scheme 1, path 2) has been reported by our group for M =
W and CS2 and benzene as solvents.4 The rate-law derived from
the mechanism, using the steady-state approximation for the
intermediate species (solvent)M(CO)5 (or M(CO)5 for unassisted
displacement), is given by eqn (4).
The reactions of [60]fullerene with piperidine under flooding
conditions ([pip] ꢁ [C60]) were monitored by observing and record-
ing the absorbance values to confirm that the increasing second
and third segments of the triphasic plots correspond to piperidine–
[60]fullerene adducts formation. Fig. 6 shows an increasing plot of
absorbance vs. time for a reaction of piperidine with [60]fullerene
in benzene at 298 K. The reactions are first-order with respect to
[60]fullerene and the rate constant values are [pip]-dependent. The
ratio of three piperidine molecules per [60]fullerene molecule was
established by elemental analysis. Further characterization of the
(NC5H10)3H3C60 adduct and kinetics studies of its formation are
underway.
−d[S]/dt = kobsd [S]
In eqn (4), S = (g -C60)M(CO)5 and
kobsd = (k1k2[L]/(k−1[C60] + k2[L]))
(4)
2
(5)
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1293–1299 | 1297
©

View Article Online
(CO)5W–C60 bond enthalpy (=131 kJ mol−1) and the enthalpy
of activation for (CO)5W–C60 bond breakage (=107 kJ mol−1) is
approximately the same as the reported TS stabilization of W(CO)5
by decalin (ca. 25 kJ mol−1, vide infra). The TS stabilization for
the analogous Cr and Mo species is approximately 3 kJ mol−1 and
13 kJ mol−1, respectively.
The negligible TS stabilization in the chromium species
[benzene–(CO)5Cr–C60] is at variance with the estimated entropy
of activation (DS1 = −35(8) J K−1 mol−1) for the benzene-
‡
assisted [60]fullerene displacement from (CO)5Cr–C60. A negative
entropy of activation value suggests an associative displacement
of [60]fullerene from the parent complex because there is a loss
of entropy in the step governed by k1 where benzene and the
parent complex become part of a single TS structure. One may
expect that a rather large TS stabilization (at least larger than for
the W and Mo analogs) should accompany a process where the
TS is highly solvated and therefore stabilized.42 Thus, a negligible
TS stabilization may suggest, either that the (CO)5Cr–benzene
interaction is rather weak or that [benzene–Cr(CO)5–C60] is a
high energy TS. This interpretation prompted the computation
of (CO)5M–benzene bond enthalpies (Table 3). The estimated TS
stabilization for W(CO)5 (ca. 24 kJ mol−1) and for Mo(CO)5 (ca.
13 kJ mol−1) seems to be a direct result of the intermediate species
stabilization ((CO)5M–benzene bond enthalpies are 76 kJ mol−1
and 66 kJ mol−1, respectively). Interestingly, for the Cr analog, the
appreciable stabilization of the intermediate species ((CO)5Cr–
benzene bond enthalpy = 57 kJ mol−1) does not produce the
expected transition state stabilization.
Scheme 1 Proposed mechanisms for [60]fullerene displacement from
(g -C60)M(CO)5. Path A describes a solvent-assisted displacement of
[60]fullerene whereas Path B describes a dissociative displacement. Ia and Ib
are steady-state intermediates. TS1 and TS2 are plausible transition states.
2
Under flooding conditions where [C60] ≈ 0, eqn (5) reduces to
eqn (6),
kobsd = (k1k2[pip]/(k−1[C60] + k2[pip])) ≈ k1
(6)
The condition that [C60] ≈ 0 is experimentally met since the
only source of [60]fullerene in the mixture comes from substrate
2
dissociation. If [60]fullerene dissociation from (g -C60)M(CO)5 is
It may be instructive to compare the (CO)5M–benzene bond
enthalpies estimated from DFT methods in this investigation
with the corresponding experimental values estimated from the
enthalpies of activation for the benzene–M bond fission in
photogenerated species in solution and in the gas phase.43–46 For
example, the benzene–Cr(CO)5 bond enthalpy (39.3(4)43 kJ mol−1
not assisted (or just barely assisted) by the solvent, then the en-
thalpies of activation values should be close to the thermodynamic
[60]fullerene–metal bond enthalpies.
Since the solvent may affect the stability of the transition
state (TS) leading to formation of the intermediate species
(solvent)M(CO)5, the feasibility to estimate M–C60 bond en-
thalpies from enthalpies of activation for the step where M–
2
and 48(5)44 kJ mol−1) reported for the photogenerated (g -
benzene)Cr(CO)5 in solution is 8 to 18 kJ mol−1 smaller than
the value reported (57(3) kJ mol−1)45 for reactions of the gas
‡
C60 ruptures (DH1 ) should depend on the degree of solvent
2
involvement in the step governed by k1 in scheme 1. Indeed,
this TS stabilization is not negligible as reported elsewhere.
For example, the difference between the gas phase (CO)5W–
phase-produced (g -benzene)Cr(CO)5 species. The benzene–Cr
bond enthalpy estimated from DFT (57 kJ mol−1) and the value
2
estimated from gas phase-produced (g -benzene)Cr(CO)5 are the
CO bond dissociation enthalpy (=192
12 kJ mol−1)37 and
same. The smaller benzene–Cr bond enthalpy estimated from
reactions of the photogenerated species in solution suggests partial
benzene–Cr bond breakage in the transition state [benzene–
Cr(CO)5]. Results from experiments using isotopically labeled
benzene (C6d6) led to the proposition of an agostic C–H–
Cr interaction in the transition state.43 Therefore the hapticity
the (CO)5W–CO bond dissociation enthalpy in decalin (=167
7 kJ mol−1)38 implies a 25 kJ mol−1 TS stabilization by decalin. This
25 kJ mol−1 TS stabilization may be a result of a C–H–W agostic
interaction.39–41 Since the agostic-W bond enthalpy is reported to
be approximately 56 kJ mol−1,42 a TS stabilization of 25 kJ mol−1
suggests a concerted CO–W bond breakage and decalin–W bond
formation with partial agostic C–H–W bond formation in the
transition state.
2
of the benzene–Cr coordination changes from g = 2 in (g -
benzene)Cr(CO)5 to g = 1 in the transition state.43 Likewise,
the benzene–Mo(CO)5 (52.0(2) kJ mol−1) and benzene–W(CO)5
(62(2) kJ mol−1) bond enthalpies for the displacement of benzene
To assess the degree of solvent involvement in the transition state
[(solvent–M(CO)5–C60], (solvent = benzene), the enthalpies of ac-
2
from the photogenerated (g -benzene)M(CO)5 species (M = Mo,
2
tivation for [60]fullerene displacement from (g -C60)M(CO)5 were
W)46 are smaller than the corresponding DFT values (101 kJ mol−1
for Mo, 131 kJ mol−1 for W), also indicating partial benzene–M
bond breakage in the transition state notwithstanding the positive
entropy of activation values reported (+17.2(8) J K−1 mol−1 for
Mo and +20(5) J K−1 mol−1 for W) for the same reactions.46
The apparent lack of transition state stabilization47 due to
solvation in the Cr analog prompted DFT calculations on the
relative (CO)5M–benzene and (CO)5M–C60 bond energies (in the
compared to thermodynamic M–C60 bond enthalpies obtained
by DFT methods. Table 3 shows the thermodynamic (CO)5M–
C60 bond enthalpies, (CO)5M–benzene bond enthalpies, enthalpies
of activation, and entropies of activation for the solvent-assisted
(CO)5M–C60 bond breakage. The results from this investigation
agree with the reported [(CO)5W–decalin] TS stabilization. For
example, the difference (ca. 24 kJ mol−1) between the gas phase
1298 | Dalton Trans., 2007, 1293–1299
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
transition state [C60–(CO)5M–benzene]; M = Cr and W). The
results, presented in Table 3, confirm that the relative ((CO)5M–
benzene bond strength/(CO)5M–C60 bond strength) is significant
in the chromium analog (12 kJ mol−1/59 kJ mol−1 = 0.20) and
16 M. D. Westmeyer, T. B. Rauchfuss and A. K. Verma, Inorg. Chem.,
1996, 35, 7140–7147.
17 E. Nakamura and M. Sawamura, Pure Appl. Chem., 2001, 73(2), 355–
359.
18 M. Sawamura, H. Iikura, T. Ohama, U. E. Hacler and E. Nakamura,
J. Organomet. Chem., 2000, 599, 32–36.
19 M. Sawamura, H. Likura and E. Nakamura, J. Am. Chem. Soc., 1996,
118, 12850–12851.
negligible in the tungsten analog (2.3 kJ mol−1/68 kJ mol−1
0.03).
=
20 H. B. Bu¨rgi, E. Blanc, D. Schwarzenbach, S. Liu, Y. Lu, M. M. Kappes
and J. A. Ibers, Angew. Chem., Int. Ed. Engl., 1992, 31, 640.
21 S. Satpathy, Chem. Phys. Lett., 1986, 130, 545.
Conclusion
22 J. M. Hawkins, A. Meyer, T. A. Lewis, S. Loren and F. J. Hollandrer,
Science, 1991, 252, 312.
23 Q. Xie, E. Pe´rez-Cordero and L. Echegoyen, J. Am. Chem. Soc., 1992,
114, 3978.
The formation of a crowded transition state [benzene–Cr(CO)5–
C60] may prevent transition state stabilization by solvation.
It has been suggested that [60]fullerene, with its high elec-
tron affinity23–29 similar to those of electron-deficient olefins48,49
and capable of accepting electron density from chromium(0),
24 Y. Ohsawa and T. Saji, J. Chem. Soc., Chem. Commun., 1992, 781.
25 L. Echegoyen and L. E. Echegoyen, Acc. Chem. Res., 1998, 31, 593.
26 D. Dubois, M. T. Jones and K. M. Kadish, J. Am. Chem. Soc., 1992,
114, 6446.
2
2
may stabilize congested transition states (as in [fac-(g -C60)(g -
dppe)(L)Cr(CO)3] (dppe = 1,2-bis(diphenylphosphino)ethane).50
However, this [60]fullerene-induced transition state stabilization
may not be operative in the present study because [60]fullerene is
27 F. Zhou, C. Jehoulet and A. J. Bard, J. Am. Chem. Soc., 1992, 114,
11004.
28 D. Dubois, K. M. Kadish, S. Flanagan, R. E. Haufiler, L. P. F. Chibante
and L. J. Wilson, J. Am. Chem. Soc., 1991, 113, 4364.
29 R. C. Haddon, L. E. Brus and K. Raghavachari, Chem. Phys. Lett.,
1986, 125, 459.
30 R. C. Haddon, Science, 1993, 261, 1545.
31 T. J. Seiders, K. K. Baldridge, J. M. O’Connor and J. S. Siegel, J. Am.
Chem. Soc., 1997, 119, 4781.
2
a departing ligand in [benzene–Cr(CO)5–C60], whereas in [fac-(g -
2
C60)(g -dppe)(L)Cr(CO)3], [60]fullerene stays in the chromium(0)
coordination sphere during the reaction.
32 J. R. Rogers and D. S. Marynick, Chem. Phys. Lett., 1993, 205, 197.
33 R. C. Haddon, J. Comput. Chem., 1998, 19, 139.
34 E. D. Jemmis and M. Manoharan, Curr. Sci., 1999, 76, 1122.
35 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03,
Gaussian, Inc.: Wallingford CT, 2004.
Acknowledgements
Acknowledgment is made to the Donors of The Petroleum
Research Fund, administered by the American Chemical Society,
(ACS-PRF # 41267-B3). Helpful comments and suggestions from
reviewers are gratefully acknowledged.
References
1 Y. Ocasio-Delgado, J. De Jesu´s-Segarra and J. E. Corte´s-Figueroa,
J. Organomet. Chem., 2005, 690, 3366.
2 L. A. Rivera-Rivera, G. Crespo-Roma´n, D. Acevedo-Acevedo, Y.
Ocasio-Delgado and J. E. Corte´s-Figueroa, Inorg. Chim. Acta, 2004,
357, 881.
3 Y. Ocasio-Delgado, L. A. Rivera-Rivera, G. Crespo-Roman and J. E.
Corte´s-Figueroa, Inorg. React. Mech., 2003, 5, 13.
4 L. A. Rivera-Rivera, F. D. Colo´n-Padilla, Y. Ocasio-Delgado, Y.
Martinez-Rivera, S. Mercado-Feliciano, C. M. Ramos and J. E. Cortes-
Figueroa, Inorg. React. Mech., 2002, 4, 49.
36 J.-I. Aihara, Theor. Chem. Acc., 1999, 102, 134–138.
37 K. E. Lewis, D. M. Golden and G. P. Smith, J. Am. Chem. Soc., 1984,
106, 3905.
5 L. Rivera, F. Colo´n-Padilla, A. Del Toro-Novale´s and J. E. Corte´s-
Figueroa, J. Coord. Chem., 2001, 54, 143.
6 L.-C. Song, P.-C. Liu, J.-T. Liu, F.-H. Su, G.-F. Wang, Q.-M. Hu, P.
Zanello, F. Laschi and M. Fontani, Eur. J. Inorg. Chem., 2003, 3201–
3210.
7 A. J. Babcock, J. Li, K. Lee and J. R. Shapley, Organometallics, 2002,
21, 3940–3946.
38 G. R. Graham and R. J. Angelici, Inorg. Chem., 1967, 6, 2082.
39 G. R. Dobson, K. J. Asali, C. D. Cate and C. W. Cate, Inorg. Chem.,
1991, 30, 4471.
40 M. Brookhart and M. L. H. Green, J. Organomet. Chem., 1983, 250,
395.
8 P. Zanello, F. Laschi, A. Cinquantini, M. Fontani, K. Tang, X. Jin and
L. Li, Eur. J. Inorg. Chem., 2000, 1345–1350.
9 P. Zanello, F. Laschi, M. Fontani, C. Mealli, A. Ienco, K. Tang, X. Jin
and L. Li, J. Chem. Soc., Dalton Trans., 1999, 965–970.
10 P. Zanello, F. Laschi, M. Fontani, L.-C. Song and Y.-H. Zhu,
J. Organomet. Chem., 2000, 593–594, 7–11.
11 L.-C. Song, J.-T. Liu, Q.-M. Hu, G.-F. Wang, P. Zanello and M.
Fontani, Organometallics, 2000, 19, 5342.
12 A. L. Balch, L. Hao and M. M. Olmstead, Angew. Chem., Int. Ed.
Engl., 1996, 35, 188.
41 J.-Y. Saillard and R. Hoffmann, J. Am. Chem. Soc., 1984, 106, 2006.
42 J. M. Morse, G. H. Parker and T. J. Burkey, Organometallics, 1989, 8,
2471.
43 S Zhang, G. R. Dobson, V. Zang, H. C. Bajaj and R. van Eldik, Inorg.
Chem., 1990, 29, 3477.
44 A. A. Bengali and T. F. Stumbaugh, Dalton Trans., 2003, 354.
45 J. R. Wells, P. G. House and E. Weitz, J. Phys. Chem., 1994, 98, 8343.
46 S. Zhang and G. R. Dobson, Inorg. Chim. Acta, 1991, 181, 103.
47 G. S. Hammond, J. Am. Chem. Soc., 1955, 77, 334.
48 E. D. Jemmis, M. Manoharan and P. K. Sharma, Organometallics,
2000, 19, 1879.
13 P. J. Fagan, J. C. Calabrese and B. Malone, Science, 1991, 252, 1160.
14 A. N. Chernega, M. L. H. Green, J. Haggitt and A. H. H. Stephens,
J. Chem. Soc., Dalton Trans., 1998, 755.
49 F. Nunzi, A. Sgamellotti, N. Re and C. Floriani, J. Chem. Soc., Dalton
Trans., 1999, 3487.
15 J. T. Park, H. Song, J. J. Cho, M. K. Chung, J. H. Lee and I. H. Suh,
Organometallics, 1998, 17, 227.
50 E. Igartua-Nieves, Y. Ocasio-Delgado and J. E. Corte´s-Figueroa,
J. Coord. Chem., 2007, DOI: 10.1080/00958970600873596.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 1293–1299 | 1299
©