Modelling catalytic turnover frequencies in ionic liquids: The determination of the bimolecular rate constant for solvent displacement from [(C6H6)Cr(CO)2Solv] in 1-w-butyl-3- methylimidazolium hexafluorophosphate

Article abstract of DOI:10.1039/b315781d

The bimolecular rate constant for solvent displacement, k₂, from [(C₆H₆)Cr(CO)₂Solv] by an incoming ligand has been determined in the room temperature ionic liquid, [bmim][PF ₆], and is compared to that for the same process in cyclohexane and dichloroethane.

Full text of DOI:10.1039/b315781d

Modelling catalytic turnover frequencies in ionic liquids: the
determination of the bimolecular rate constant for solvent displacement
from [(C6H6)Cr(CO)2Solv] in 1-n-butyl-3-methylimidazolium
hexafluorophosphate†
Konrad Swiderski,a Andrew McLean,*a Charles M. Gordon*b and D. H. Vaughana
a Department of Chemistry & Chemical Engineering, University of Paisley, Paisley, Scotland, UK PA1 2BE.
E-mail: mcle-ch0@paisley.ac.uk
b Institut für Technische Chemie und Makromolekulare Chemie, RWTH Aachen, Worringer Weg 1, 52074
Aachen, Germany. E-mail: gordon@itmc.rwth-aachen.de
Received (in Cambridge, UK) 4th December 2003, Accepted 21st January 2004
First published as an Advance Article on the web 3rd February 2004
The bimolecular rate constant for solvent displacement, k2,
from [(C6H6)Cr(CO)2Solv] by an incoming ligand has been
determined in the room temperature ionic liquid, [bmim][PF6],
and is compared to that for the same process in cyclohexane and
dichloroethane.
Variation of the concentration of acetonitrile gave k2 = 5.0 3 107
mol21 dm3 s21 for the displacement reaction at 25 °C. This value
may be compared to k2 = 1 3 107 mol21 dm3 s21 for displacement
of cyclohexane by CO obtained by other workers.2
The transient spectra in C2H4Cl2 and CH3OH show weak
depletions at 340 and 320 nm and isosbestic points at 370 and 350
nm respectively. Neither solvent was displaced by acetonitrile at
concentrations up to 5.0 mol dm23 on our time-scales ( < 200 ms),
suggesting that k2 5 104 mol21 dm3 s21 in these cases.
The time dependent spectral behaviour in [bmim][PF6] was more
complex than in conventional organic solvents (Fig. 2). The
spectrum recorded after 20 ns shows depletion in the region around
340 nm and an isosbestic point at 370–380 nm. The species
responsible for this spectrum decayed via first order kinetics
Room temperature ionic liquids (RTILs) are excellent media in
which to carry out transition metal based catalysis.1 However, there
is a lack of quantitative information concerning reaction rate
constants and mechanisms in such systems as compared to
conventional solvents. There are, of course, several steps in any
catalytic cycle that may be influenced by the choice of solvent, any
one of which may be rate determining. The obvious starting point
for any detailed study investigating RTIL effects on catalytic cycles
is to establish the relative ability of a reactant to displace the
coordinated RTIL from a transition metal site compared to
conventional solvent systems. As specifically noted in a recent
review,1e no such rate constants have been determined to date. We
report here the first value for a bimolecular rate constant for
displacement of an RTIL from a transition metal complex and
compare our result to those in conventional solvent systems.
We chose [(C6H6)Cr(CO)2Solv] as our model system because it
can be readily generated by UV photolysis of [(C6H6)Cr(CO)3].2
The photo-induced decarbonylation of [(C6H6)Cr(CO)3] in alkane
solution yielding a solvent bound intermediate, [(C6H6)Cr(CO)2-
Solv], is well established:2
(1)
[(C6H6)Cr(CO)3] + Solv ? [(C6H6)Cr(CO)2Solv] + CO
Solvents that are bound weakly in [(C6H6)Cr(CO)2Solv] can be
displaced via a bimolecular exchange mechanism2 with a rate
constant, k2:
Fig. 1 Transient difference spectrum in cyclohexane, [acetonitrile] = 0.023
mol dm23. Inset (a); Transient at 350 nm. Inset (b); Plot of kobs versus
(2)
[(C6H6)Cr(CO)2Solv] + L ? [(C6H6)Cr(CO)2L] + Solv
[acetonitrile] at 25 °C, k2 = 5.0 3 107 mol21 dm3 s21
.
We therefore carried out a series of laser flash photolysis
experiments (LFP)3 using 355 nm excitation on [(C6H6)Cr(CO)3]
in cyclohexane, dichloroethane, methanol and the RTILs
[bmim][PF6] and [bmim][(CF3SO2)2N] at 25 °C.4 Time resolved
spectra were determined from point to point under N2 saturated
conditions and k2 obtained from plots of the observed first order
rate constant for the decay of [(C6H6)Cr(CO)2Solv] transients, kobs
,
as a function of [L], where L = acetonitrile. The results are
summarized in Figs. 1 and 2.
The spectrum in cyclohexane is shown in Fig. 1. The transient
spectrum recorded 500 ns after the laser pulse is consistent with that
previously reported for [(C6H6)Cr(CO)2(C6H12)], with a strong
depletion at 340 nm and an isosbestic point at 380 nm.2 In the
presence of acetonitrile reaction (2) occurs, resulting in formation
of [(C6H6)Cr(CO)2(CH3CN)], the spectrum of which can be seen 5
ms after the laser pulse in Fig. 1. This assignment was confirmed by
the immediate formation of a spectrum identical to that recorded at
5 ms when [(C6H6)Cr(CO)3] was irradiated in neat acetonitrile.
Fig. 2 Transient difference spectrum in [bmim][PF6]. Inset (a); Transient at
370 nm. Inset (b); Plot of kobs versus [acetonitrile] at 25 °C, k2 = 5.8 3 106
† Electronic supplementary information (ESI) available: synthesis of
RTILs. See http://www.rsc.org/suppdata/cc/b3/b315781d/
mol21 dm3 s21
.
590
C h e m . C o m m u n . , 2 0 0 4 , 5 9 0 – 5 9 1
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

forming a second species with an isosbestic point at 310 nm. The
first order rate constant for this process, kobs, measured at 370 nm,
was 5 3 106 s21. We repeated the experiment with water-saturated
[bmim][PF6] (ca. 1.5% w/w H2O),5 which corresponds to [H2O] =
1.6 M. An identical product spectrum was observed to grow in more
rapidly (kobs = 8 3 106 s21), which strongly suggests that the final
product is [(C6H6)Cr(CO)2(H2O)]. From these data we can estimate
a bimolecular rate constant k2 ≈ 5 3 106 mol21 dm3 s21 for the
reaction of [(C6H6)Cr(CO)2(RTIL)] with water. These results
clearly show that, despite care being taken to dry the [bmim][PF6]
prior to use and minimal subsequent exposure of the samples to the
air, a relatively large amount of water was nevertheless absorbed.
Finally, we investigated the reaction of [(C6H6)Cr(CO)2(RTIL)]
with varying concentrations of acetonitrile in [bmim][PF6]. A value
of k2 = 5.8 3 106 mol21 dm3 s21was determined for displacement
of [bmim][PF6], as shown in inset (b) of Fig. 2. It should be noted
that this value is around fifty times smaller than that for a diffusion
controlled reaction in this solvent.6
In principle, both [bmim]+ and [PF6]2 ions may be initially
bound to the [(C6H6)Cr(CO)2] intermediate generated immediately
after the laser pulse. However, a blue-shifted difference spectrum
was obtained on changing the RTIL anion to [(CF3SO2)2N]2 with
an isosbestic point and depeletion at 350 and 320 nm respectively.
Furthermore, we could not displace [bmim][(CF3SO2)2N] from
[(C6H6)Cr(CO)2Solv] at an acetonitrile concentration of 5 mol
dm23 which indicates that a more strongly bound [(C6H6)Cr(CO)2-
Solv] species was now present. Taken together, these observations
suggest that binding of the [PF6]2 anion is responsible for the
spectral signature at 20 ns in Fig. 2.
This is a model catalytic system designed purely to ascertain the
relative ability of a ligand to displace an RTIL from [(C6H6)-
Cr(CO)2Solv] compared to weakly coordinated conventional
solvents. Obviously these results are most pertinent to situations
where solvent displacement by a reactant is the rate determining
step in the catalytic cycle. It is therefore extremely satisfying to
note that enhanced turn over frequencies (TOFs) have been
observed for the cationic Ni catalysed oligomerisation of ethene to
higher olefins in [bmim][PF6] compared to that in dichloromethane
and butane-1,4-diol (TOF
= 12712, 1852 and <
10 h21
respectively).7 The enhanced TOF in the RTIL was considered to
result from a combination of weak binding of [PF6]2 to the catalytic
site and decreased product solubility in this solvent. Our present
studies clearly support the former conclusion and also highlight the
importance of controlling the water content of RTILs noted by the
authors.8
In conclusion, we have unambiguously demonstrated that
[bmim][PF6] may be displaced more easily from a ‘vacant’
transition metal binding site of an organometallic complex than
even a ‘low polarity’ solvent such as C2H4Cl2 and furthermore, that
the ease of displacement can be tuned by variation of the RTIL
anion. These initial results do not correlate with ligand basicity as
measured by b values of these solvents in a simple manner.9 We are
continuing our studies into the mechanism of solvent displacement
and dependence of k2 on the Lewis basicity of the anion, and will
present a more detailed report in the near future.
We would like to thank the University of Paisley for funding a
PhD Studentship (KS).
The isosbestic points between [(C6H6)Cr(CO)2Solv] and
[(C6H6)Cr(CO)3] do not show great variation with solvent.
However, we can arrive at the approximate spectrochemical series:
C6H12 ~ [bmim][PF6] < C2H4Cl2 < CH3OH < H2O based on
isosbestic points of 380, 370–380, 370, 350 and 300 nm
respectively. In an earlier study using a Cu complex which acts as
a measure of solvent basicity, it was found that the strength of
[PF6]2 co-ordination to Cu was comparable to or slightly weaker
than dichloromethane.7 Although not overwhelming, the spectro-
scopic evidence therefore supports the following order of Cr–Solv
interaction strength: C6H12 < [bmim][PF6] < C2H4Cl2.
Notes and references
1 (a) J. D. Holbrey and K. R. Seddon, Clean Prod. Proc. I, 1999, 223; (b)
T. Welton, Chem. Rev., 1999, 2071; (c) P. Wasserscheid and W. Keim,
Angew. Chem., Int. Ed., 2000, 39, 3772; (d) C. M. Gordon, Appl. Catal.,
A: Gen., 2001, 222, 101; (e) D. Zhao, M. Wu, Y. Kou and M. Enze, Catal.
Today, 2002, 74, 157.
2 B. S. Creaven, M. W. George, A. C. Ginzburg, C. Hughes, J. M. Kelly,
I. M. McGrath and M. T. Pryce, Organometallics, 1993, 12, 3127.
3 For description of LFP setup see: A. C. Benniston, P. R. MacKie, L. J.
Farrugia, G. Smith, S. J. Teat and A. J. McLean, New J. Chem., 2001, 4,
58.
4 For preparation of the ionic liquid see: J. G. Huddleston, H. D. Willauer,
R. P. Swatlowski, A. E. Visser and R. D. Rogers, Chem. Commun., 1998,
1765; P. Bonhote, A. Das, N. Papageorgiou, K. Kalanasundram and M.
Gratzel, Inorg. Chem., 1996, 35, 1168.
The evidence from solvent displacement kinetics is far less
ambiguous; we find that k2 values for solvent displacement from
[(C6H6)Cr(CO)2Solv] by acetonitrile follow the trend: C6H12
>
[bmim][PF6] > C2H4Cl2. Our results show that the rate constant for
displacement of [PF6]2 from [(C6H6)Cr(CO)2Solv] by acetonitrile
is up to two orders of magnitude greater than that of the next
strongest binding solvent in our series, C2H4Cl2. Furthermore, it
should be reiterated that despite the ‘hydrophobic’ nature of
[bmim][PF6], significant concentrations of H2O can be present
even after careful drying. The result of this is that high
concentrations of displacing ligand are required in order to ensure
effective competition with displacement by H2O. Thus, water
content might represent a problem in the application of RTILs as
solvents in some catalytic systems.
5 N. V. K. Sudhir, J. F. Brennecke and A. Samanta, Chem. Commun., 2001,
413.
6 A. J. McLean, M. J. Muldoon, C. M. Gordon and I. R. Dunkin, Chem.
Commun., 2002, 1880.
7 P. Wasserscheid, C. M. Gordon, C. Hilgers, M. J. Muldoon and I. R.
Dunkin, Chem. Commun., 2001, 1186.
8 The TOF for the Ni system is much lower in [bmim][(CF3SO2)2N] than
[bmim][PF6], which is in agreement with our observations regarding ease
of solvent displacement by acetonitrile. P. Wasserscheid (personal
communication).
9 L. Crowhurst, P. Mawdsley, J. M. Perez-Arlandis, P. A. Slater and T.
Welton, Phys. Chem. Chem. Phys., 2003, 5, 2790.
C h e m . C o m m u n . , 2 0 0 4 , 5 9 0 – 5 9 1
591