How important is the release-return mechanism in olefin metathesis?

Article abstract of DOI:10.1002/chem.201001832

From whose bourns no traveller returns, puzzles the will: UV/Vis, ¹⁹F NMR, and fluorescence spectroscopic studies (see graphic) provide no evidence supporting the boomerang mechanism in Grubbs-Hoveyda complexes, which is the return of the isopropoxy styrene as a benzylidene ether ligand following its release during catalyst initiation.

Full text of DOI:10.1002/chem.201001832

DOI: 10.1002/chem.201001832
How Important Is the Release–Return Mechanism in Olefin Metathesis?
Tim Vorfalt, Klaus J. Wannowius, Vasco Thiel, and Herbert Plenio*[a]
Hoveyda[1] and later Blechert[2] first reported on stable
ruthenacarbenes derived from Grubbs II complexes, in
which benzylidene and PCy3 are replaced by a bidentate
benzylidene ether ligand.[3] Such complexes combine excel-
lent stability with remarkable catalytic activity in various
types of olefin metathesis reactions.[3b,4] It was claimed, that
the initiation step involves dissociation of the benzylidene
ether and that following the olefin metathesis reaction, the
bidentate isopropoxy styrene returns to the ruthenium as a
benzylidene ligand; this was termed boomerang or release–
return mechanism.[1,5] Based on the observation of deuterat-
ed and non-deuterated benzylidene ether ligand crossover
between bead-immobilized Grubbs–Hoveyda complexes,
Hoveyda et al. provided evidence that the initially proposed
release–return mechanism is operative.[6] Even though this
mechanism has been recognized by numerous others with
various Grubbs–Hoveyda-type olefin metathesis catalysts,[5,7]
the important question of whether this mechanism contrib-
utes substantially or only marginally has not been re-
solved.[8] Recently, Grela et al. reported on exchange experi-
ments between deuterated and non-deuterated benzylidene
ether ligand and styrene ether. First the background reac-
tion between styrene ether and deuterated benzylidene
ether ligand in complexes 3 and 4 was probed. During ring-
closing metathesis (RCM) reactions the deuterated and non-
deuterated labels were equilibrated much faster than the
background reaction. Based on this, it was concluded that
the whole amount of precatalyst applied was involved in the
catalytic reaction and was then regenerated by the release–
return mechanism.[9]
We and others recently discovered that tagging of ligands
in metal complexes with fluorescent dyes can provide useful
information on ligand dissociation reactions due to changes
in the fluorescence intensity during the catalytic reactions.[10]
Transition-metal ions often quench the fluorescence of such
dyes and consequently fluorophore-tagged ligands coordi-
nated to such metals, render weakly fluorescent com-
plexes.[11] However, upon dissociation of the tagged ligand
from a metal complex, the fluorescence is restored due to
the spatial separation between the two components. When
the dissociation of such a tagged ligand represents a key
step in the catalytic cycle, the monitoring of the fluorescence
intensity will provide detailed mechanistic information. This
could thus become a key experiment for probing the re-
lease–return mechanism.
We want to report here on the release of the benzylidene
ether ligand under real catalytic conditions during olefin
metathesis reactions using Hoveyda-type precatalysts and
(more importantly) whether a return of this ligand occurs to
a significant extent. To enable the monitoring of such reac-
tions, in addition to the normal Grubbs–Hoveyda complex
3, closely related complexes with different spectroscopic re-
porter groups were needed. The fluorine substituted com-
plex 1 was obtained according to a known procedure;[12] the
new fluorophore-tagged complex 2 was synthesized in good
yields using the respective dansyl-tagged styrene.
It was important, to first show that the dansyl fluorophore
and the fluorine groups located para to the ether oxygen do
not disturb the RCM activity of complexes 1 and 2 to a sig-
nificant degree. With a view to the Hammett constants sp-
[a] Dr. T. Vorfalt, Dr. K. J. Wannowius, Dipl.-Chem. V. Thiel,
Prof. Dr. H. Plenio
Organometallic Chemistry, TU Darmstadt
Petersenstrasse 18, 64287 Darmstadt (Germany)
E-mail: plenio@tu-darmstadt.de
AHCTUNTGRENN(GUN NHSO2Me)=0.03 and sp(F)=0.06 this appears to be un-
likely.[13] RCM reactions for the conversion of DEDAM (di-
ethyl diallyl malonate) catalyzed by the tagged complexes 1
and 2 and Grubbs–Hoveyda complex 3 were monitored and
the respective conversion-time curves found to be compara-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001832. It contains a descrip-
tion of the synthetic work, characterization data for the new com-
pounds, fluorescence and UV/Vis traces and NMR spectra.
12312
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12312 – 12315

COMMUNICATION
ble (Figure S1 in the Supporting Information), with complex
3 showing the slowest and 1 the fastest conversion. Appa-
rently, neither the dansyl nor the fluorine tag interferes with
the olefin metathesis reaction to a really significant extent.
Next the fluorescence of complex 2 was monitored in tol-
uene, in the absence of RCM substrate, to obtain informa-
tion on the stability of 2 under the experimental conditions
(blank experiment). The initial fluorescence intensity is
weak and only a very slow increase with time was observed
(Figure 1, trace a). This experiment was repeated with
nature of the substrate and not caused by a separate decom-
position reaction independent from substrate conversion. It
is also evident that fluorophore-tagged ligands can be highly
useful tools for mechanistic studies in transition-metal catal-
ysis. The RCM reaction with DEDAM and catalyzed by 3
was also monitored by UV/Vis spectrometry. The UV/Vis
spectra of 1 and 2 are characterized by a distinct absorbance
at 380 nm. This absorbance disappears during the RCM re-
action, but does not return after the RCM reaction. There-
fore an independent UV/Vis experiment provides no evi-
dence for a return of the isopropoxy styrene.
To obtain a better picture of the RCM reaction, the fluo-
rescence experiments were repeated for different olefin con-
centrations, at constant concentrations of complex 2 (5.3ꢁ
10ꢀ5 m). In contrast to UV/Vis experiments, providing a
metal-centered view of olefin metathesis event, the fluores-
cence experiments furnish information on the (liberated)
fluorophore, which is formed after the first olefin metathesis
reaction. For an analysis of the initiation reaction the same
kinetic model as reported before is used.[16] Accordingly, the
Grubbs–Hoveyda complex is first activated by the olefinic
RCM substrate. The activated complex then allows substrate
molecules to react to the product. The derived rate expres-
sion for the fluorescence intensity I for conversion of the
Figure 1. Fluorescence evolution of a toluene solution of complex 2
(trace a, blank) and during the RCM of DEDAM (trace b).
substrate is: I=(I0ꢀI1)/
ACHTNUGTNERN(NUG 1+kobs t)+I1. The fitting of the
added RCM substrate (DEDAM, 0.5 mol% of 2).[14] Within
a few minutes a strong increase of the fluorescence intensity
occurred, until after about 120 min a plateau was reached
and held for the next 18 h (Figure 1, trace b, only the first
10 h are shown). To probe whether this fluorescence intensi-
ty corresponds to the liberation of all dansyl-tagged benzyli-
dene ether, the dansyl-tagged complex 2 was reacted with
1000, 2500, and 5000 equivalents of ethyl vinyl ether. All of
those reactions lead to the same final fluorescence intensity,
corresponding to quantitative initiation and full liberation of
the fluorophore (Figure S3 in the Supporting Informa-
tion).[15]
The weak initial fluorescence of a solution of complex 2 is
indicative of efficient fluorescence quenching. The initiation
of the olefin metathesis reaction leads to the dissociation of
the fluorophore tag and thus to the spatial separation of
ruthenium and the fluorophore. Consequently, the fluores-
cence of the dansyl group is restored. However, a release–
return mechanism also requires that the liberated fluoro-
phore-tagged styrene returns to the ruthenium. Under the
conditions employed here, this should result in the partial
quenching of the fluorescence. However, the fluorescence
intensity remains virtually constant after the RCM reaction.
The same type of fluorescence–time curve was observed for
two additional RCM reactions with N, N-diallylcarbamate
and N, N-tosyldiallylamide carried out under the same con-
ditions (Figures S9 and S10 in the Supporting Information).
For the three RCM reactions with different substrates cata-
lyzed by pre-catalyst 2 the full fluorescence evolution can be
slower or faster, due to different rates of the initiation reac-
tions for the three olefinic substrates. This also indicates
that the increase of fluorescence intensity is related to the
fluorescence–time curves for the DEDAM reactions yields
the respective kobs, and a linear fit of the various kobs versus
substrate concentration provides the second-order rate con-
stant for catalyst initiation k1 =(15.0ꢁ2)ꢁ10ꢀ3 mꢀ1 sꢀ1 of 2.
This rate constant for the initiation is close to the initiation
rate obtained from UV/Vis experiments k1 =(23.8ꢁ3ꢁ
10ꢀ3 mꢀ1 sꢀ1) for complex 3,[16] which shows that the fluores-
cence and the UV/Vis experiments report on the same
event. For the same reaction of the fluorine-tagged complex
1 a faster k1 =(57.5ꢁ2)ꢁ10ꢀ3 mꢀ1 sꢀ1 was obtained from UV/
Vis experiments.
When using UV/Vis and fluorescence spectroscopy with
complexes 2 and 3 it is difficult to obtain precise data on the
identity of the species formed in the course of olefin meta-
thesis reactions. The 19F NMR signals in 1 and other fluo-
rine-containing derivatives can provide such information
and consequently the evolution of the 19F NMR signal in the
RCM reaction of DEDAM with 1 was recorded. The excel-
lent sensitivity of 19F NMR spectroscopy allows the perfor-
mance of these experiments under the same conditions as
before. Initially the 19F signal of complex
1
(d=
ꢀ126.2 ppm) was observed, but the ongoing initiation reac-
tion leads to a single new signal (d=ꢀ125.4 ppm), which
corresponds to that of the free 3-fluoro-6-isopropoxy styrene
and which is also the only 19F NMR signal observed at the
end of the RCM reaction. Again there is no evidence sup-
portive of a release–return mechanism.
This finally leads us to conclude, that under the conditions
of the catalytic reaction, the return of the styrene ether to
the ruthenium to reform the Grubbs–Hoveyda type complex
does not occur to a significant extent. Nonetheless, we
cannot ignore a significant number of publications, which
Chem. Eur. J. 2010, 16, 12312 – 12315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12313

H. Plenio et al.
report on the re-isolation of Grubbs–Hoveyda complex after
olefin metathesis reactions. Two scenarios are conceivable to
explain this. First we take an exemplary look at a represen-
tative procedure for such a RCM reaction. We note that up
to 5 mol% of the pre-catalyst 3 were used, which is much
more than needed for such a conversion; as it is now known
from the work of Dorta et al.[17] and from our own work[16]
that Grubbs–Hoveyda type catalysts can be extremely active
for RCM reactions at high concentrations of olefinic sub-
strates. In the reaction discussed, a 0.1m substrate concen-
tration was utilized, which is relatively high. Secondly, the
reaction workup took place after only 10 min reaction time.
This time is too short to activate a significant proportion of
the pre-catalyst. A consequence of this is that the initiation
reaction is slower than the olefin metathesis conversion.
This becomes evident in Figure 2; even at a 1 mol% loading
complexes studied (1, 2, and 3). This conclusion is mainly
based on three observations: 1) The highly characteristic
380 nm UV/Vis absorbance in complex 1 and 3 disappears
in the course of the catalyst initiation and is not restored
after the olefin metathesis reaction; 2) the dansyl fluoro-
phore in complex 2 is released during the initiation period,
the bright fluorescence is turned on and does not disappear
after the RCM reaction; and 3) the characteristic 19F NMR
signal of the fluorine-tagged isopropoxy styrene from com-
plex 1 grows in during catalyst initiation and is the only fluo-
rine-containing species following olefin metathesis. We thus
believe that the absence of return is much more common
than previously thought. It is likely, that the re-isolation of
the Grubbs–Hoveyda complex following olefin metathesis
reactions is primarily caused by incomplete activation of the
initial Grubbs–Hoveyda complex. This happens when the
catalytic transformation is faster than catalyst initiation and
occurs when much more catalyst complex is used than is re-
quired for a certain olefin metathesis reaction.[18] For low
catalyst loading (just sufficient to effect the desired transfor-
mation) most of the precatalyst will undergo activation. This
active olefin metathesis catalyst does not live forever and
should fall victim to various decomposition reactions after a
certain number of turnovers.[19]
Acknowledgements
This work was supported by the DFG, grant Pl 178/8-3. We wish to thank
Dr. A. Marx (Merck KgaA) for recording the 19F NMR spectra.
Figure 2. Fluorescence-time curves for the RCM reaction of DEDAM
with complex 2 at 0.2 and 1 mol% loading.
Keywords: fluorescence · olefin metathesis · release–return
mechanism · ruthenium · UV/Vis spectroscopy
only a small amount of catalyst is activated after 30 min.
With a 5 mol% loading and 10 min reaction time, the pro-
portion of activated catalyst will be very small. Consequent-
ly, the large majority of Grubbs–Hoveyda complex will
remain unchanged and available for re-isolation. Even at
1 mol% loading (Figure 2) less than half of the complex is
activated after 60 min, while at 0.2 mol% loading after
60 min activation is not quantitative.
As an alternative scenario, the concentration of the reac-
tion mixture during the evaporation of the solvent might
lead to the re-association of the isopropoxy styrene. We
tested this by repeating the RCM reaction with DEDAM
and 0.2 mol% of complex 3 under the conditions of the
fluorescence experiment. After 6 h reaction time (quantita-
tive initiation, Figure 2) the volatiles were evaporated. The
remaining oil was dissolved in a minimum amount of
CH2Cl2 and the solution applied to thin-layer chromato-
graphic plates. However, the green spot indicative of the
Grubbs–Hoveyda complex was not observed.
[1] a) J. S. Kingsbury, J. P. A. Harrity, P. J. Bonitatebus, A. H. Hoveyda,
J. Am. Chem. Soc. 1999, 121, 791–799; b) S. B. Garber, J. S. Kings-
bury, B. L. Gray, A. H. Hoveyda, J. Am. Chem. Soc. 2000, 122,
8168–8179.
[2] S. Gessler, S. Randl, S. Blechert, Tetrahedron Lett. 2000, 41, 9973–
9976.
[3] a) C. E. Diesendruck, E. Tzur, N. G. Lemcoff, Eur. J. Inorg. Chem.
2009, 4185–4203; b) C. Samojłowicz, M. Bieniek, K. Grela, Chem.
Rev. 2009, 109, 3708–3742; c) G. C. Vougioukalakis, R. H. Grubbs,
Chem. Rev. 2010, 110, 1746–1787.
[4] a) A. H. Hoveyda, A. R. Zhugralin, Nature 2007, 450, 243–251;
b) A. Michrowska, R. Bujok, S. Harutyunyan, V. Sashuk, G. Dolgo-
nos, K. Grela, J. Am. Chem. Soc. 2004, 126, 9318; c) A. H. Hoveyda,
D. G. Gillingham, J. J. V. Veldhuizen, O. Kataoka, S. B. Garber, J. S.
Kingsbury, J. P. A. Harrity, Org. Biomol. Chem. 2004, 2, 8–23;
d) K. M. Kuhn, J.-B. Bourg, C. K. Chung, S. C. Virgil, R. H. Grubbs,
J. Am. Chem. Soc. 2009, 131, 5313–5320.
[5] M. Ahmed, A. G. M. Barrett, D. C. Braddock, S. M. Cramp, P. A.
Procopiou, Tetrahedron Lett. 1999, 40, 8657–8662.
[6] J. S. Kingsbury, A. H. Hoveyda, J. Am. Chem. Soc. 2005, 127, 4510–
4517.
[7] a) M. Ahmed, T. Arnauld, A. G. M. Barrett, D. C. Braddock, P. A.
Procopiou, Synlett 2000, 1007–1009; b) L. Jafarpour, M.-P. Heck, C.
Baylon, H. M. Lee, C. Mioskowski, S. P. Nolan, Organometallics
2002, 21, 671–679; c) J. Dowden, J. Savovꢂc, Chem. Commun. 2001,
37–38; d) C. Hongfa, H.-L. Su, H. S. Bazzi, D. E. Bergbreiter, Org.
In conclusion, neither fluorescence spectroscopy, UV/Vis
experiments nor 19F NMR spectroscopy provide evidence
supportive of a significant contribution of a release–return
mechanism in RCM reactions in the Grubbs–Hoveyda-type
12314
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12312 – 12315

Release–Return Mechanism
COMMUNICATION
Lett. 2009, 11, 665–667; e) M. Sꢃßner, H. Plenio, Angew. Chem.
2005, 117, 7045–7048; Angew. Chem. Int. Ed. 2005, 44, 6885–6888;
f) S. Varray, R. Lazaro, J. Martinez, F. Lamaty, Organometallics
2003, 22, 2426–2435; g) Q. Yao, M. Sheets, J. Organomet. Chem.
2005, 690, 3577–3584; h) D. Rix, F. Caijo, I. Laurent, L. Gulajski, K.
Grela, M. Mauduit, Chem. Commun. 2007, 3771–3773; i) S. W.
Chen, J. H. Kim, C. E. Song, S. G. Lee, Org. Lett. 2007, 9, 3845–
3848; j) H. Clavier, F. Caijo, E. Borrꢄ, D. Rix, F. Boeda, S. P. Nolan,
M. Mauduit, Eur. J. Org. Chem. 2009, 4254–4265.
[12] M. Zaja, S. J. Connon, A. M. Dunne, M. Rivard, N. Buschmann, J.
Jiricek, S. Blechert, Tetrahedron 2003, 59, 6545–6558.
[13] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195.
[14] This experiment was done immediately after the first experiment
under precisely the same experimental conditions as the blank ex-
periment, in order to enable comparision of absolute fluorescence
intensities.
[15] It was not possible to precisely correlate the fluorescence intensity
with the amount of liberated fluorophore as the fluorescence inten-
sity is changed by the presence of ruthenium complex.
[16] T. Vorfalt, K. J. Wannowius, H. Plenio, Angew. Chem. 2010, 122,
5665–5668; Angew. Chem. Int. Ed. 2010, 49, 5533–5536.
[17] M. Gatti, L. Vieille-Petit, X. Luan, R. Mariz, E. Drinkel, A. Linden,
R. Dorta, J. Am. Chem. Soc. 2009, 131, 9498–9499.
[18] With a large of amount of precatalyst and a comparatively low
amount of active species, decomposition reactions originating from
the active species will be less relevant.
[8] J. J. Van Veldhuizen, D. G. Gillingham, S. B. Garber, O. Kataoka,
A. H. Hoveyda, J. Am. Chem. Soc. 2003, 125, 12502–12508.
[9] M. Bieniek, A. Michrowska, D. L. Usanov, K. Grela, Chem. Eur. J.
2008, 14, 806–818.
[10] a) V. Sashuk, D. Schoeps, H. Plenio, Chem. Commun. 2009, 770–
772; b) J.-H. Sohn, K. H. Kim, H.-Y. Lee, Z. S. No, H. Ihee, J. Am.
Chem. Soc. 2008, 130, 16506–16507; c) S. M. Canham, J. Y. Bass, O.
Navarro, S.-G. Lim, N. Das, S. A. Blum, Organometallics 2008, 27,
2172–2175; d) A. Kiel, J. Kovacs, A. Mokhir, R. Krꢅmer, D.-P.
Herten, Angew. Chem. 2007, 119, 3427–3430; Angew. Chem. Int.
Ed. 2007, 46, 3363–3366; e) V. Sashuk, L. H. Peeck, H. Plenio,
Chem. Eur. J. 2010, 16, 3983–3993.
[19] S. H. Hong, A. G. Wenzel, T. T. Salguero, M. W. Day, R. H. Grubbs,
J. Am. Chem. Soc. 2007, 129, 7961–7968.
[11] R. Krꢅmer, Angew. Chem. 1998, 110, 804–806; Angew. Chem. Int.
Ed. 1998, 37, 772–773.
Received: June 29, 2010
Published online: September 13, 2010
Chem. Eur. J. 2010, 16, 12312 – 12315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12315