High turnover numbers with ruthenium-based metathesis catalysts

Article abstract of DOI:10.1002/1615-4169(200208)344:6/7<671::AID-ADSC671>3.0.CO;2-G

Effective turnover numbers (TON's) and turnover frequencies (TOF's) of unprecedented magnitude have been obtained for the metathesis of various substrates in the presence of ruthenium-based catalysts, without the use of an additional solvent. For the self

Full text of DOI:10.1002/1615-4169(200208)344:6/7<671::AID-ADSC671>3.0.CO;2-G

FULL PAPERS
High Turnover Numbers with Ruthenium-Based Metathesis
Catalysts
Maarten B. Dinger, Johannes C. Mol*
Institute of Molecular Chemistry, Faculty of Science, Universiteit van Amsterdam, Nieuwe Achtergracht 166, 1018 WV
Amsterdam, The Netherlands
Fax: ( ‡ 31)-20-5256456, e-mail: jcmol@science.uva.nl
Received: January 28, 2002; Accepted: April 18, 2002
Abstract: Effective turnover numbers (TON×s) and gave TON×s 6 times higher (TON >640,000) and
turnover frequencies (TOF×s) of unprecedented mag- initial TOF×s 20 times higher than those of 2a at
nitude have been obtained for the metathesis of ambient temperature. At elevated temperatures,
various substrates in the presence of ruthenium-based TOF×s were twice as high and initial TOF×s were
catalysts, without the use of an additional solvent. For 6 times as high for 2b relative to 2a, and with nearly
the self-metathesis of 1-octene, the 2nd generation 100% selectivity. Initial TOF×s exceeding 3,800 sÀ
1
ˆ
Grubbs× catalyst (IMesH2)(PCy3)(Cl)2Ru CHPh (2a) could be achieved with 2b at 608C. Under similar
conditions, comparable TON×s for the self-metathesis
dene) gave TON×s 5 times higher than those of the 1st of trans-4-decene and methyl oleate, and the ring-
(IMesH2 ˆ 1,3-dimesityl-4,5-dihydroimidazol-2-yli-
ˆ
generation catalyst (PCy3)2(Cl)2Ru CHPh (1) while closing metathesis of diethyl diallylmalonate were
maintaining a high degree of selectivity for the also obtained.
formation of the product 7-tetradecene. Moreover,
ˆ
the catalyst (IPrH2)(PCy3)(Cl)2Ru CHPh (2b), bear-
ing the extremely bulky 1,3-bis(2,6-diisopropylphen- Keywords: catalytic activity; N-heterocyclic carbene
yl)-4,5-dihydroimidazol-2-ylidene ligand (IPrH2), (NHC) ligands; metathesis; P ligands; ruthenium
Introduction
these systems disadvantageous for use in large-scale
reactions. While these points are especially pertinent for
The last few years have seen a significant resurgence in potential industrial applications for these catalysts, they
interest for olefin metathesis chemistry.[1] The renewed are also relevant when relatively inexpensive substrates
popularity can be largely attributed to the introduction are to be reacted. We have therefore focused our
of the well-defined, functional-group tolerant homoge- attention on the possibilities of utilizing catalysts 1 and
neous ruthenium-derived catalyst (PCy3)2(Cl)2 2a for metathesis with the goal of maximizing the
ˆ
Ru CHPh (1) developed by Grubbs and cowork- effective TON, a figure that is, surprisingly, rarely
ers,[1a,1f] together with the molybdenum-based catalysts reported for homogeneous metathesis catalysts.
developed by Schrock.[1b,1e] Especially the ruthenium-
derived catalyst systems have garnered much popularity
N
N
N
N
PCy3
with organic chemists for their ease of use (high air and
moisture stability) and exceptional tolerance toward
reactive functional groups. The presence of functional
groups in substrates often thwarts traditional metathesis
catalysts.[2] With the recent advent of the much more
active 2nd generation ruthenium catalyst (IMesH2)
Cl
Cl
Cl
Ru
Ru
Ru
Cl
Cl
Cl
Ph
Ph
Ph
PCy3
PCy3
PCy3
1
2a
2b
ˆ
(PCy3)(Cl)2Ru CHPh (IMesH2 ˆ 1,3-dimesityl-4,5-di-
We chose 1-octene as a suitable substrate for initial
hydroimidazol-2-ylidene) (2a), incorporating a bulky study, since it is inexpensive and widely available. We
neutral N-heterocyclic carbene (NHC) ligand,[3] the now report results that reveal that, in fact, very low
scope of olefin metathesis in organic synthesis has catalyst loadings (<1 ppm) can be successfully used in
become even greater.
the self-metathesis of 1-octene, Equation (1).
Although catalysts 1and 2a are powerful tools in organic
synthetic strategies, the relatively high cost of rutheni-
um, coupled with the relatively high catalyst loading
generally employed (typically 0.1 5 mol %),[4] render
[Ru]
+
2
ꢀ1†
5
5
Adv. Synth. Catal. 2002, 344, No. 6+7
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1615-4150/02/3446+7-671 677 $ 17.50+.50/0
671

Maarten B. Dinger, Johannes C. Mol
FULL PAPERS
This success led us to also examine the metathesis of
Complex 2b largely resembles its mesityl bearing
several other types of substrate, e.g., the self-metathesis analogue, with the exception of the 31P NMR spectrum
of trans-4-decene, Equation (2), and methyl oleate, giving the PCy3 signal slightly (2.4 ppm) up-field relative
Equation (3), as well as the ring-closing metathesis of to that of 2a at 28.1 ppm. No free PCy3 was observed.
diethyl diallylmalonate, Equation (4).
Like 2a, complex 2b is brownish in appearance, and the
1H NMR spectrum shows the diagnostic low-field
benzylidene proton resonance at 19.77 ppm; 2a shows
the corresponding signal at 19.59 ppm. In contrast to the
report by F¸rstner,[6] we did not observe the formation
of the dihydride complex (PCy3)2Ru(H)2(Cl)2 as a by-
product in our (similar) synthetic procedure, and clean
material could be directly isolated from the crude
reaction residue by addition of methanol. However,
we did notice that in pure samples of 2b in C6D6, using
standard (not airtight) NMR tubes, two additional
singlet signals gradually appeared in the benzylidene
region of the 1H NMR spectrum at 18.95 and 17.82 ppm
in a ꢁ 1:4 ratio. Full exposure to air led to a more rapid
loss (minutes) of the benzylidene signal of the parent
compound concomitant with growth of the new signals
and the formation of tricyclohexylphosphine oxide
(evidenced in the 31P NMR spectrum). No other
31P NMR signals were observed. Furthermore, addition
of excess Cu(I)Cl directly to a C6D6 solution of 2b gave
identical spectra to the air-exposed sample, but without
2
[Ru]
ꢀ2†
ꢀ3†
ꢀ4†
+
O
2
O
[Ru]
O
O
+
O
7
7
7
7
O
EtOOC
COOEt
EtOOC
COOEt
[Ru]
+
ˆ
O PCy3 formation. Addition of PCy3 to the Cu(I)Cl
treated sample regenerates 2b in good yield. From these
experiments, we can tentatively conclude that the
identity of at least one of these benzylidene-containing
The synthesis, characterization and reactivity of the derivatives of 2b is the 14-electron phosphine-dissoci-
slightly modified 2nd generation Grubbs× catalyst 2b is ated species proposed in the catalytic cycle of olefin
also reported, this compound giving metathesis reac- metathesis.[8,9] Noteworthy is that similar observations
tivity differing significantly from that of either of the have recently been noted for (IMesH2)(PCy3)(I)2
1
ˆ
standard ruthenium catalysts 1 and 2a.
Ru CHPh, which in the H NMR showed new benzy-
lidene signals at similar positions, 18.14 and 17.17 ppm.[8]
Attempts at the isolation of the phosphine-free deriv-
ative of 2b from the Cu(I)Cl reaction, which appears to
be quite stable in solution and the solid state, have been
hampered by its high solubility in common organic
solvents.
Results and Discussion
Preparation and Properties of Catalyst 2b
In addition to exploring TON×s for catalyst 2a, we
wondered if a further increase in the steric bulk of the Metathesis of 1-Octene
NHC ligand might impart additional reactivity and,
perhaps more importantly, longevity to the active Maximum Effective Turnover Number
catalyst. Because steric effects play an important role
in ruthenium-based metathesis catalysts,[5] we prepared Like all metathesis reactions, the metathesis of 1-octene
the 1,3-bis(2,6-diisopropylphenyl)-4,5-dihydroimidazo- is fully reversible. However, when the product ethene is
line (IPrH2) analogue of 2a. The desired complex (2b), allowed to escape, the reaction is rendered essentially
has also recently been prepared and characterized by irreversible, Equation (1). Metathesis catalysts also
F¸rstner and coworkers,[6] although its catalytic activity bring about the non-productive metathesis of the
has not yet been reported. We reasoned that the product 7-tetradecene. Because these metathesis events
significant increase in steric bulk of IPrH2 relative to cannot be followed, the total turnover number for a
IMesH2, together with its superior electron-donating given catalyst for the metathesis of 1-octene cannot be
properties relative to the well-studied unsaturated calculated with any degree of certainty. Moreover,
analogue, IPr,[7] might give rise to a highly active because the active metathesis catalyst is the phosphine
metathesis initiator.[8]
dissociated species,[8] whose concentration at any given
672
Adv. Synth. Catal. 2002, 344, 671 677

High Turnover Numbers with Ruthenium-Based Metathesis Catalysts
FULL PAPERS
Trace amounts of dodecene arising from the self-meta-
thesis of 2-octene were also detected. The data clearly
show that the selectivity is dependent on reaction time,
temperature and concentration. For example, in the case
of catalyst 1, when all metathesis activity had ceased, the
remaining unreacted 1-octene was slowly isomerized
giving almost equal parts of 1-, 2-, and 3-octene. This
finding suggests that the decomposed metathesis initia-
tor becomes a quite efficient double-bond isomerization
catalyst. Therefore, if metathesis active species are still
present late in the reaction, significant amounts of
secondary products can be formed. While this process
limits the potential recyclability of the remaining
unreacted substrate, the judicious choice of catalyst
loading, reaction time, and reaction temperature can
certainly control, if not eliminate, the isomerization
Figure 1. Effective TONand total conversion as a function of
the number of equivalents of 1-octene for catalyst 2a at 228C.
time cannot yet be accurately determined, the actual reaction.
number of substrate molecules that have undergone
metathesis by a particular ruthenium center also cannot
be determined. Thus, the only meaningfulfigure that can Effect of Temperature
be calculated with any degree of certainty is the effective
TON, this being the total number of 1-octene molecules Next, metathesis experiments ranging from ambient
converted to metathesis products per molecule of the conditions (228C) up to the boiling point of octene
catalyst precursor.
(1228C) were carried out to determine the effect of
When the maximum capabilities of the catalyst are temperature on TON×s for the catalysts 1, 2a, and 2b
reached, the effective TONshould remain constant over
a wide concentration range if the levels of impurities in
(Figure 2).
At room temperature (228C), catalyst 1 gives a
the 1-octene that can poison or decompose the catalyst maximum TONof 21,000. The 2nd generation catalysts,
are sufficiently low. Accordingly, the quality of the 1- in particular 2b, performed significantly better. As
octene used for the various experiments was verified by already mentioned above, catalyst 2a gave an optimal
testing the initiator 2a over a wide range of dilutions, viz. TONof 95,000, while the TONachieved with catalyst 2b
a ratio of 9,600 386,000 equivalents of 1-octene to exceeded 640,000. The reason for the appreciably higher
catalyst at 228C. Figure 1 graphically depicts the results TONof the catalyst 2b relative to 2a is not entirely clear,
of this study.
but could conceivably be attributed to a faster initiation
The maximum effective TONfor catalyst 2a is rate (phosphine dissociation) together with a lower
reached when about 95,000 molecules of substrate occurrence of non-productive metathesis of 7-tetrade-
have been converted. This maximum TONcan be
cene (vide infra) due to the increased steric bulk of the
achieved only when >145,000 equivalents of 1-octene IPrH2 ligand relative to IMesH2. Faster initiation rates
are used, giving a <65% conversion. Additional 1- should give rise to higher TON×s, which has recently
octene leads to lower conversions, but essentially the been demonstrated for 2a relative to the faster initiating
same TON×s, while less 1-octene results in a higher triphenylphosphine analogue, (IMesH2)(PPh3)(Cl)2
conversion but overall a more inefficient use of the
catalyst.
All subsequent metathesis experiments were carried
out with an appropriate excess of substrate, to ensure the
highest possible TONwas reached in each case.
Selectivity
The selectivity for the formation of 7-tetradecene was
generally very high ( > 99%), but we did notice slightly
lower selectivities ( ꢁ 95%) for the more concentrated
mixtures. The major secondary product was determined
Figure 2. Effective TONas a function of temperature for
catalysts 1, 2a, and 2b for the metathesis of 1-octene.
Reactions were deemed complete when no further meta-
by GC/MS to be tridecene, most probably formed from
the cross-metathesis of 1-octene and 2-octene, the latter
ˆ
formed from C C bond isomerization of the former. thesis activity was observed.
Adv. Synth. Catal. 2002, 344, 671 677
673

Maarten B. Dinger, Johannes C. Mol
FULL PAPERS
ˆ
Ru CHPh, the latter allowing much lower catalyst
loadings.[8] The increased steric bulk of the IPrH2 ligand
might also impart additional protection of the ruthe-
nium center, thereby slowing down the substrate
independent decomposition pathways, although only
very little is currently known about the formation and
identity of the exhausted catalyst.
At higher temperatures, divergent results are ob-
served. Catalyst 1 shows a steady increase in TONat
elevated temperatures, up to 28,500 at 608C. Above
808C, a sharp drop-off in activity was noted; this
presumably represents the decomposition temperature
of the catalyst. Catalyst 2a does not succumb appreci-
ably to thermally-induced degradation until over 1008C.
From this point on a slight, but steady, decrease in TON
was observed, ending with ꢁ 50% reduction in TONat
1228C. Catalyst 2b was found to be more thermally
Figure 3. Effective TONas a function of time for catalysts 1,
2a, and 2b in the metathesis of 1-octene at 608C, using 67,400,
718,000, and 1,154,000 equivalents of substrate, respectively.
sensitive than 2a, and shows an almost linear decrease in time. As expected, catalyst 2a gave a significantly higher
activity at temperatures higher than 508C. That said, maximum observed TOF of 4.2 sÀ at 45 min reaction
1
even at 1208C, the TONof 2b is still higher than the time. Noteworthy is that an initiation period of ꢁ 30 min
maximum value recorded for 2a.
was observed for 2a. In contrast, catalyst 2b displayed no
Interestingly, the catalyst 2a displays a very sudden detectable initiationÀperiod, and a TOF more than
1
increase in TONbetween 50 and 52 8C (Figure 2). Over 20 times higher (88 s ) than that of 2a was measured
this narrow temperature range the effective TON after 15 minutes. Consequently, catalyst 2b was com-
reproducibly doubled to ꢁ 300,000, and thereafter pletely exhausted after 7 h, while 2a still showed some
remained approximately constant up to ꢁ 1008C. We activity after 15 h.
also noticed a sharp decrease of 10 15% in selectivity
Figure 3 shows the rates of metathesis at 608C. At this
above 508C after prolonged reaction times. Secondary higher temperature, 2b again displays much higher
metathesis products detected in these cases were non- activity than both 1 and 2a; a TOF of 3,800 sÀ was
1
ene, decene, dodecene, and tridecene, with only the measured after 1 min reaction time for 2b, while 1 and 2a
latter detected at any significant level at temperatures showed maximum TOF×s of 60 and 670 s-1, respectively.
lower than 508C. Possibly, ꢁ 508C represents a thresh-
old temperature for the more rapid initiation of 2a,
which then leads to higher TON×s.[8] This phenomenon Stereoselectivity
was not observed for 2b, which maintained a relatively
constant activity over a wide temperature range while Finally, we investigated the stereochemistry of the 7-
retaining a very high degree of selectivity. The reason for tetradecene isolated from 1-octene self-metathesis con-
the different behavior of the two 2nd generation ducted at 22 and 608C. Because the cis and trans isomers
catalysts at elevated temperatures is not entirely clear, of 7-tetradecene could not be readily separated by gas
but possibly involves the sterically induced faster chromatography, the cis:trans ratio of the metathesis
initiation rate of 2b relative to 2a.
product was analyzed for the three catalysts by quanti-
tative 13C NMR experiments.[10] Table 1 shows a sum-
mary of the data.
Effective Turnover Frequency
The results are consistent with cis:trans ratios deter-
mined for a variety of reaction products involving
We also followed the rates of the metathesis reaction for
the three catalysts under investigation at 22 and 608C.
As for the determination of the TON, the unknown
number of active catalytic species present in the solution
means that the turnover frequency (TOF) of a particular
molecule of catalyst cannot be determined. Thus, when
measuring the rate profiles of the reactions, the values at
any given time represent a TONaveraged over all of the
ruthenium species (those that are catalytically active,
decomposed, and dormant) present in solution.
Table 1. cis:trans ratios of the 7-tetradecene formed from the
metathesis of 1-octene by catalysts 1, 2a, and 2b, at 22 and
608C.
Temperature [8C]
Catalyst
cis:trans ratio
22
1
40:60
16:84
16:84
40:60
22:78
22:78
2a
2b
1
2a
2b
60
At 228C, catalyst 1 showed a low activity, and gave a
maximum effective TOF of 2.4 sÀ after 5 min reaction
1
674
Adv. Synth. Catal. 2002, 344, 671 677

High Turnover Numbers with Ruthenium-Based Metathesis Catalysts
FULL PAPERS
Table 2. Self-metathesis of 1-octene, trans-4-decene and methyl oleate, and ring-closing metathesis of diethyl diallylmalonate
at 558C in the presence of catalysts 1, 2a, and 2b.[a] Neat solutions.
Substrate
1-Octene
Catalyst Equivalents of Substrate Total Conversion [%] Selectivity [%] Effective TON
1
104,000
490,000
1,400,000
4,200
27
60
41
30
48
15
10
45
25
16
17
25
> 98
91
28,500
295,000
630,000
1,250
2a
2b
1
2a
2b
1
2a
2b
1
2a
2b
> 98
> 98
> 98
> 98
> 98
91[d]
> 98
> 98
> 98
> 98
trans-4-Decene
1,200,000
208,000
25,000
990,000
240,000
50,000
570,000
31,000
2,500
440,000
46,000
8,000
Methyl oleate[b]
Diethyl diallylmalonate[c]
1,160,000
960,000
200,000
240,000
[a]
Reaction time 24 h. Reactions were carried out over a wide catalyst concentration range. Highest values are given.
[b] Successful metathesis dependent on the batch of methyl oleate used; TON×s > 100,000 could be reproducibly obtained with
2a.
[c]
TON×s and selectivities were calculated using an internal standard (nonane).
Slightly lower TON×s but with > 98% selectivity could be obtained at 228C.
[d]
aliphatic olefin substrates metathesized by catalysts 1 and methyl oleate, catalyst 2a gives, unexpectedly,
and 2a.[4b,4d,11] The differing steric influence of the NHC TON×s higher than those measured for 1-octene, while
ligands in catalysts 2a and 2b does not contribute to the catalyst 2b gives significantly lower values. We speculate
stereoselectivity of the catalysts, since identical data that the bulkier IPrH2 ligand in 2b hinders the approach
were obtained for the two systems.
of the internal olefin molecules to the Ru center,
resulting in lower TOF×s and consequently lower
TON×s. Studies to give a more definitive insight into
this somewhat surprising result are in progress.
Metathesis of Other Substrates
For the RCM of diethyl diallylmalonate, high TON×s
The high TON×s described above for the metathesis of 1- could be obtained without any added solvent using
octene led us to examine three other substrates: trans-4- catalyst 2a.[13] This result sharply contrasts that of the
decene, methyl oleate, and diethyl diallylmalonate, seminal report by Grubbs, where a minimal catalyst
Equations (2) (4). Methyl oleate and diethyl diallyl- loading of 0.05 mol % (2,000 equivalents of substrate)
malonate are of particular interest since they contain of 2a was required to initiate the RCM of diethyl
ester groups that can decompose many traditional diallylmalonate using CH2Cl2 as solvent.[3] The im-
metathesis catalysts.[2] Table 2 summarizes the results proved efficacy of 2b relative to 2a for 1-octene meta-
obtained using the three initiators under investigation thesis does not fully translate to the ring-closing meta-
for both the self-metathesis of trans-4-decene, Equa- thesis of diethyl diallymalonate, although 2b did con-
tion (2), and methyl oleate, Equation (3), as well as the sistently give slightly improved ( ꢁ 20%) TON×s relative
ring-closing metathesis (RCM) of diethyl diallylmalo- to 2a.
nate, Equation (4).
From the data it is clear that the very high TON×s
shown for catalysts 2a and 2b can be readily extended to Conclusions
include other substrates, notably those containing ester
groups. Catalyst 1, however, shows only moderate The TON×s for the 1st generation Grubbs× catalyst 1 and
TON×s for the other substrates. Especially, a clear the 2nd generation catalysts 2a and 2b have been
drop-off in activity on going from terminal olefins to determined covering a wide temperature range (22
internal ones for catalyst 1 is evident. The result 1228C) for the metathesis of 1-octene. The metathesis
obtained for methyl oleate is consistent with the TON reactions examined proceeded smoothly in the absence
previously reported, while that for trans-4-decene was of added solvent, this being especially noteworthy given
found to be much higher.[12]
current environmental concerns regarding solvent
As has been observed by F¸rstner and coworkers,[6] use.[14] The 2nd generation catalysts display much higher
different NHC ligands present in the 2nd generation TON×s than the 1st generation catalyst. Similarly high
metathesis catalysts can give quite different metathesis TON×s could also be obtained for the metathesis of
activity, depending on the substrate, and this was also the trans-4-decene, methyl oleate and diethyl diallylmalo-
case with catalysts 2a and 2b. For both trans-4-decene nate. These high TON×s are, to the best of our knowl-
Adv. Synth. Catal. 2002, 344, 671 677
675

Maarten B. Dinger, Johannes C. Mol
FULL PAPERS
vacuum applied to the flask. The octene was then deoxygen-
ated by a series of degassing (by evacuation of the flask),
followed by filling with nitrogen. The required amount of
octene was then immediately transferred by a N2-flushed
syringe to degassed Schlenk flasks in which the metathesis
experiments were to be carried out. NMR spectra were
recorded on a Varian Mercury 300 spectrometer, at 300.14,
75.48, and 121.50 MHz for the proton, carbon and phosphorus
channels, respectively.
edge, unprecedented for ruthenium-based metathesis
catalysts. A maximum TONof 110,000 has been
reported for the poorly defined catalyst produced in
situ from a RuBr3 ¥ x H2O/PCy3/2-butyne-1,4-diol diac-
etate/H2 mixture.[15]
The results reported here strongly imply that the
ruthenium-based catalysts are, in all likelihood, often
being used well below their maximum capabilities in
many metathesis reactions, with loadings of 0.1
5 mol % (20 1,000 equivalents of substrate) being the
norm.[4] While from the present work it is clear that
solvents are not required for metathesis, they are
unavoidable for RCM of large rings (to circumvent
undesirable intermolecular metathesis reactions),[16]
and also for solid substrates. It is quite likely that the
presence of relatively low levels of catalyst poisons in
the substrate play an important role in catalyst decom-
position when very low catalyst loadings are used.
Minimization of this problem should be controllable by
more careful purification of the substrate. Furthermore,
catalyst 2a shows an unanticipated performance jump at
ꢁ 508C for the metathesis of 1-octene, which was not
observed for catalysts 1 and 2b. Given that metathesis
reactions using ruthenium catalysts are generally con-
ducted in refluxing CH2Cl2, better results might be
anticipated when using 2a if a higher boiling solvent
coupled with a slightly higher reaction temperature is
used.
The selectivity of the metathesis reaction was found to
be highly dependent on the reaction conditions, but was
generally excellent for all three catalysts, independent of
the substrates tested. While turnover frequencies were
only determined at two temperatures (22 and 608C), it is
noteworthy that the rate of metathesis was found to be
extremely rapid at 608C for the 2nd generation catalysts.
We are currently exploring additional reaction con-
ditions to further optimize the effective TONfor the
metathesis of 1-octene and other substrates, together
with investigations of additional 2nd generation
Grubbs× catalysts bearing modified NHC ligands.
Complex 2b
Potassium tert-pentoxide solution ( ꢁ 1.7 M in toluene, 190 mL,
0.323 mmol) was added to a suspension of 1,3-bis(2,6-diiso-
propylphenyl)-4,5-dihydroimidazolinium chloride (0.135 g,
0.316 mmol) in hexanes (20 mL). The stirred mixture was
subsequently placed in an oil bath at 508C and reacted for
5 min; the milky suspension rapidly became cloudy. The
solution was then added (by way of a stainless steel cannula
fitted with a filter) to catalyst 1 (0.200 g, 0.243 mmol)
suspended in hexanes (20 mL). The combined solution was
heated by oil bath at 508C for 30 min, resulting in a clear dark
brown solution. The solvent was completely removed under
vacuum and degassed methanol (20 mL) was added to the dark
brown residue, producing a fine powder upon stirring (1 h).
The solid was filtered under N2, washed with methanol (4 Â
15 mL), and finally dried under vacuum to give 2b as a light
brown powder; yield: 0.175 g (80%). 31P{1H} NMR (C6D6): d ˆ
1
ˆ
28.1; H N MR (CD6): d ˆ 19.77 (s, 1H, Ru CHPh), 8.25 (br s,
6
1H, ortho CH), 7.23 7.12 (multiple peaks, 5H, aryl CH), 6.97
(t, 2H, para CH, 3JHH ˆ 7.8 Hz), 6.70 (s br, 2H, aryl CH), 4.19
3
(sept, 1H, CH(CH3)2, JH,H ˆ 6.5 Hz), 3.76 (t, 2H, CH2CH2,
3JH,H ˆ 7.8 Hz), 3.69 (t, 2H, CH2CH2, 3JH,H ˆ 7.8 Hz), 2.27 [s br,
1H, CH(CH3)2], 1.70 1.02 [multiple peaks, 45H, PCy3 and
13
1
2
ˆ
CH(CH3)2]; C{ H} NMR: d ˆ 296.7 (d, Ru CHPh, JC,P
ˆ
8 Hz), 222.24 [d, Ru-C(N)2, 2JC,P ˆ 80 Hz], 151.5, 149.9, 147.3,
138.5, 136.3, 131.9, 130.6, 129.3, 124.8, 124.3, 55.2 (CH2CH2),
54.6 (CH2CH2), 32.6, 32.4, 29.9, 29.6, 28.5, 28.3, 28.2, 27.8, 27.1,
26.8, 24.2, 24.1; anal. calcd. for C52H77N2Cl2PRu: C 66.93, H
8.32, N3.00%; found: C 66.84, H 8.44, N2.94%.
Metathesis Experiments
In a typical experiment, between 10 and 60 mL (64 382 mmol)
of 1-octene were used. Degassed octadecane (ꢁ5% of the
volume of octene used), which is less volatile than any of the
metathesis products, was used as an internal standard. To the
solution was then added an accurately weighed sample (six-
figure analytical balance) of solid catalyst, in the range of 0.25
2.0 mg. No additional solvents were added. In each case, the
reaction vessel was allowed to vent to an oil bubbler. A
constant slow stream of nitrogen ensured anaerobic conditions
for the duration of the experiment. All reactions were
thoroughly stirred by way of a magnetic stirrer bar. For the
various temperature experiments, the reaction vessel was
immersed in an oil bath and allowed to equilibrate to the
desired temperature before introduction of the catalyst. The
progress of the metathesis reactions was measured by GC/FID
(Carlo Erba 8000 Top) equipped with a DB-5 (J&W Scientific)
column. An error of Æ 5% was estimated for the analysis of a
Experimental Section
General Remarks
All manipulations were carried out under an inert atmosphere
of nitrogen on a vacuum line using standard Schlenk tech-
niques. All solvents used were dried and distilled under
nitrogen prior to use. The catalysts 1[17] and 2a,[3] and 1,3-
bis(2,6-diisopropylphenyl)-4,5-dihydroimidazolinium chlor-
ide[18] were prepared following literature procedures. Potassi-
um tert-pentoxide solution (1.7 M in toluene, Fluka) and
octadecane (Sigma) were used as received. 1-Octene (98%,
Aldrich) was passed through a column (20 cm  1.5 cm) of
neutral alumina (Acros, 50 200 mm), using 15 g of alumina per
100 mL of 1-octene. The column was attached to a Schlenk
flask and elution of the octene was facilitated by use of a slight
676
Adv. Synth. Catal. 2002, 344, 671 677

High Turnover Numbers with Ruthenium-Based Metathesis Catalysts
FULL PAPERS
given metathesis reaction based on the reproducibility of data
from duplicate and triplicate experiments. At higher temper-
atures ( > 808C), severe effervescence arising from ethene
production was often encountered, this unavoidably causing
undesirable and uncontrollable cooling of the reaction mix-
tures and evaporation of the substrate; errors for these
experiments are estimated at Æ 5 10%.
hedron Lett. 2001, 42, 171 174; d) F. C. Engelhardt, M. J.
Schmitt, R. E. Taylor, Org. Lett. 2001, 3, 2209 2212;
e) T.-L. Choi, A. K. Chatterjee, R. H. Grubbs, Angew.
Chem. Int. Ed. 2001, 40, 1277 1279; f) J. Louie, R. H.
Grubbs, Angew. Chem. Int. Ed. 2001, 40, 247 249;
g) A. K. Chatterjee, J. P. Morgan, M. Scholl, R. H.
Grubbs, J. Am. Chem. Soc. 2000, 122, 3783 3784.
For the metathesis of trans-4-decene (Fluka), methyl oleate
[5] L. Jafarpour, S. P. Nolan, J. Organomet. Chem. 2001,
617 618, 17 27.
[6] A. F¸rstner, L. Ackermann, B. Gabor, R. Goddard,
C. W. Lehmann, R. Mynott, F. Stelzer, O. R. Thiel,
Chem. Eur. J. 2001, 7, 3236 3253.
[7] a) L. Jafarpour, E. D. Stevens, S. P. Nolan, J. Organomet.
Chem. 2000, 606, 49 54; b) L. Jafarpour, H-J. Schanz,
E. D. Stevens, S. P. Nolan, Organometallics 1999, 18,
5414 5416; c) L. Jafarpour, J. Huang, E. D. Stevens, S. P.
Nolan, Organometallics 1999, 18, 3760 3763.
[8] M. S. Sanford, J. A. Love, R. H. Grubbs, J. Am. Chem.
Soc. 2001, 123, 6543 6554.
(Aldrich), and diethyl diallylmalonate (Aldrich), the sub-
strates were first passed through small columns of alumina
ending with a plug of celite, and thoroughly degassed. Volumes
of 1 mL of substrate were used for the reactions. Solutions of
known concentration of the catalyst were prepared in 5 mL of
hexanes for catalysts 2a and 2b, and 2 mL of benzene for
catalyst 1, and 10 50 mL quantities of these solutions were
then introduced via N2-flushed syringe to the preheated (558C)
substrates. Final reaction compositions were determined in the
same way as for 1-octene.
[9] M. S. Sanford, L. M. Henling, M. W. Day, R. H. Grubbs,
Angew. Chem. Int. Ed. 2000, 39, 3451 3453.
[10] Applicability of quantitative 13C NMR for the determi-
nation of the stereochemistry of internal olefins was
verified by integration of the olefinic region of the
13C NMR spectra obtained for mixtures of cis- and trans-
3-heptene (both commercially available) of known
composition. For 7-tetradecene, chemical shift values
are 130.0 and 130.5 ppm for the cis and trans olefinic
carbons, respectively.
[11] C. W. Lee, R. H. Grubbs, Org. Lett. 2000, 2, 2145 2147.
[12] W. Buchowicz, J. C. Mol, J. Mol. Cat. A: Chem. 1999, 148,
97 103.
[13] That no intermolecular metathesis occurred between the
substrates was verified by use of an internal standard.
[14] A. D. Curzons, D. J. C. Constable, D. N. Mortimer, V. L.
Cunningham, Green Chem. 2001, 3, 1 6.
[15] P. O. Nubel, C. L. Hunt, J. Mol. Cat. A: Chem. 1999, 145,
323 327.
[16] a) C. W. Lee, R. H. Grubbs, J. Org. Chem. 2001, 66,
7155 7158; b) C. W. Lee, R. H. Grubbs, Org. Lett. 2000,
2, 2145 2147.
[17] P. Schwab, M. B. France, J. W. Ziller, R. H. Grubbs,
Angew. Chem. Int. Ed. Engl. 1995, 34, 2039 2042.
[18] A. J. Arduengo III, R. Krafczyk, R. Schmutzler, Tetra-
hedron 1999, 55, 14523 14534.
Acknowledgements
We wish to thank SASOL for funding this research.
References and Notes
[1] For recent reviews, see: a) T. M. Trnka, R. H. Grubbs,
Acc. Chem. Res. 2001, 34, 18 29; b) R. R. Schrock, J.
Chem. Soc. Dalton Trans. 2001, 2541 2550; c) A. F¸rst-
ner, Angew. Chem. Int. Ed. 2000, 39, 3012 3043; d) L.
Yet, Chem. Rev. 2000, 100, 2963 3007; e) R. R. Schrock,
Tetrahedron 1999, 55, 8141 8153; f) R. H. Grubbs, S.
Chang, Tetrahedron 1998, 54, 4413 4450; g) S. K. Arm-
strong, J. Chem. Soc. Perkin Trans. 1 1998, 371 388;
h) C. Pariya, K. N. Jayaprakash, A. Sarkar, Coord. Chem.
Rev. 1998, 168, 1 48; i) M. Schuster, S. Blechert, Angew.
Chem. Int. Ed. Engl. 1997, 36, 2036 2056.
[2] K. J. Ivin, J. C. Mol, Olefin Metathesis and Metathesis
Polymerization, Academic Press, San Diego, London,
1997.
[3] M. Scholl, S. Ding, C. W. Lee, Grubbs, R. H. Org. Lett.
1999, 1, 953 956.
[4] For some recent examples, see: a) K. Grela, M. Bieniek,
Tetrahedron Lett. 2001, 42, 6425 6428; b) C. Pietraszuk,
H. Fischer, M. Kujawa, B. Marciniec, Tetrahedron Lett.
2001, 42, 1175 1178; c) J. A. Smulik, S. T. Diver, Tetra-
Adv. Synth. Catal. 2002, 344, 671 677
677