Unexpected results of a turnover number (TON) study utilising ruthenium-based olefin metathesis catalysts

Article abstract of DOI:10.1002/adsc.200505053

A turnover number (TON) study of ruthenium-based metathesis catalysts has been conducted for ring-closing metathesis (RCM) in dilute solution. Unexpectedly the results indicate that 1^st generation metathesis catalysts can give higher TON in RCM

Full text of DOI:10.1002/adsc.200505053

FULL PAPERS
Unexpected Results of a Turnover Number (TON) Study
Utilising Ruthenium-Based Olefin Metathesis Catalysts
Simon Maechling, Mirko Zaja, Siegfried Blechert*
Insitüt für Chemie, Fakultät II, Technische Universität Berlin, Strasse des 17. Juni 135, 10623 Berlin, Germany
Fax: (þ49)-30-3142-3619, e-mail: blechert@chem.tu-berlin.de
Received: January 26, 2005; Accepted: April 29, 2005
Abstract: A turnover number (TON) study of ruthe- generation Hoveyda–Grubbs catalyst showed unex-
nium-based metathesis catalysts has been conducted pectedly high activity, particularly when compared
for ring-closing metathesis (RCM) in dilute solution. to the 2nd generation catalysts.
Unexpectedly the results indicate that 1st generation
metathesis catalysts can give higher TON in RCM
of simple unsubstituted terminal olefins than their Keywords: alkenes; homogeneous catalysis; metathe-
second generation counterparts. In particular, the 1st sis; ring-closing metathesis (RCM); ruthenium
Introduction
represent an uneconomic use of the catalyst systems.
With emphasis on ꢀgreenerꢁ chemistry it is desirable to
With the advent ofefficient catalysts, the olefin metathe- utilise reduced catalyst loadings in order to reduce costs
sis reaction has emerged as one of the most powerful car- for industrial processes.
bon-carbon bond forming methods in organic synthesis.
Of the many transformations involving olefin meta-
This is primarily due to the disclosure of the Schrock mo- thesis, most work has concentrated on the use of 1 and
lybdenum-based catalyst,[1] followed by the ruthenium- 2, due to their commercial availability. Surprisingly, to
based catalyst 1 by Grubbs and co-workers in the mid the best of our knowledge no detailed reports exist com-
1990s,[2] which combined remarkable functional group paring the activity of these (pre)catalysts to 3 and 4, due
tolerance and general stability. The replacement of in part to the same 14-electron intermediate being re-
one of the phosphine donors with a stronger s-donating sponsible for the metathesis activity of 1 and 3, 2 and
carbene ligand, in the belief that electron-deficient in- 4. Intriguingly there have also been contradictory re-
termediates would be better stabilised, led to the 2nd gen-
eration ruthenium benzylidene complex 2. In a system-
atic study of catalysts with the general formula L1L2X2
¼
Ru CHR it was found that all ligands coordinated to
the metal centre can individually and greatly influence
the catalytic performance of a particular system.[3] It
was shown that the 2nd generation catalysts (i.e., 2), ex-
hibit improved stability/functional group tolerance and
the ability to efficiently form tri- and tetrasubstituted
double bonds. Further developments led to the seren-
dipitous discovery of (pre)catalyst 3,[4] and later the
phosphine-free alkylidene 4 bearing a chelating isopro-
poxybenzylidene ligand.[5] These (pre)catalysts display
good recyclability properties and, in particular, 4 has
been shown to perform particularly well in certain cross
metathesis (CM) reactions of electron-deficient al-
kenes, where catalyst decomposition is an important is-
sue.
The commercially available ruthenium-derived meta-
thesis catalysts (Figure 1) have been used extensively in
organic synthesis but the high cost and the high catalyst
loadings that are typically used, generally 1–10 mol %, Figure 1. Ruthenium-based olefin metathesis precatalysts.
Adv. Synth. Catal. 2005, 347, 1413–1422
DOI: 10.1002/adsc.200505053
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1413

Simon Maechling et al.
FULL PAPERS
ports in quantifying catalytic activity and efficiency of
ruthenium-based olefin metathesis (pre)catalysts.[6]
Almost all previously published studies base the activ-
ity of (pre)catalysts on initiation rates with a few stand-
ard metathesis reactions.[7] As a counterpoint to these in-
vestigations we have focused on attempts to maximise
the turnover number (TON) of a given RCM catalyst.
In particular, we were interested to what extent the iso-
propoxystyrene ligand present in 3 and 4 affect TONs
when compared to their phosphine-containing ana-
logues 1 and 2. The TON describes the degree of activity
of a catalyst; and the catalyst efficiency is described
though its turnover frequency (TOF in hꢀ1). Like all
metathesis reactions, RCM is a reversible process. How-
ever when the by-product of the metathesis reaction
(here ethene) is allowed to escape, the reaction is ren-
dered essentially irreversible. These metathesis events
cannot be followed and therefore the total turnover
number for an individual catalyst molecule cannot be
determined with any given accuracy. Therefore the
TON represents the average number of substrate mole-
cules converted into the cyclised product per molecule
of (pre)catalyst.[8]
Figure 2. RCM of 5: Conversion as a function of (pre)catalyst
loading.
thesis reactions, and the conversions were monitored by
Studies by Grubbs and co-workers have shown that HPLC. Substrate/catalyst mole ratio (S/C mole ratio) of
different phosphine-containing analogues of complexes between 5,000 and 10,000 [0.02–0.001 mol % (pre)cata-
1 and 2 affect the rate of initiation,[9] and studies by Ble- lyst] were chosen with the conversions being determined
chert and co-workers have uncovered basic rules for by HPLC, and the TON determined.
substitution of the isopropoxybenzylidene ligand to ef-
From the first results (Figure 2) it is clear that even
fect initiation in (pre)catalysts of type 3 and 4.[10] We when relatively small amounts of (pre)catalyst are
were particularly interested in the effect the substitution used, high conversions can be obtained. When compar-
of this type of ligand by others would have on the TON, ing the conversions obtained from the different (pre)ca-
in particular we were interested in whether the activity talysts utilising identical loadings it is clear that (pre)ca-
of the catalyst could be influenced by steric or electronic talyst 3 delivers higher conversions at lower catalyst
effects incorporated into the chelating isopropoxystyr- loadings in the RCM of 5 to 6.
ene ligand.
For example only 0.01 mol % of 3 is required to effect
A point that is commonly overlooked is that the fast- a 75% conversion whereas at least twice this (pre)cata-
est initiating (pre)catalyst does not necessarily afford the lyst loading is required to achieve the same conversion
highest chemical yield of a particular transformation with the other (pre)catalysts. Furthermore, at a catalyst
with the lowest catalyst loading,[11] of course, the lifetime loading of 0.006 mol %, (pre)catalyst 3 delivers a con-
of the propagating species plays an important role. We version of 60% whilst the other (pre)catalysts require
were interested in investigating the lower catalyst limit a loading of at least twice this to achieve the same conver-
to effect a ring closing metathesis transformation, to de- sion.
termine reactivity profiles and to provide a method for
The TONs that were obtained during the RCM of 5
quantifying activity for the catalysts most commonly are shown in Figure 3. The activity of the catalysts fol-
used in olefin metathesis reactions.
lows the trend 3>4>1>2. The trend is somewhat sur-
prising since previous reports have shown 2 to display
higher activity, however most previous studies have
been based on efficiency or TOF. One trend already rec-
ognised by Grubbs et al., is that catalyst activity is often
Results and Discussion
We first examined the activities of the catalysts 1, 2, 3 contrary to catalyst efficiency. Indeed, it is interesting to
and 4 in RCM reactions of simple terminal alkenes. In note that the order of catalyst efficiency has previously
order to obtain meaningful comparisons specific condi- been measured as 1ꢁ2>4ꢂ3.[10] What is clear is that
tions were developed. Initial studies were concerned higher TONs are obtained with the (pre)catalysts con-
with the RCM of 5, to yield the 5-membered ring prod- taining the chelating isopropoxystyrene ligand.
uct 6. The specific conditions employed (0.05 M in
It is thought that the same catalytically active 14-elec-
CH2Cl2, N2 atmosphere, reflux, 14 h) were chosen to re- tron species of 1 and 3, 2 and 4 are operating and that the
flect the conditions most commonly used in olefin meta- different activities of the (pre)catalysts can be reasoned
1414
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1413–1422

TON Study Utilising Ruthenium-Based Olefin Metathesis Catalysts
FULL PAPERS
astheir unsubstituted counterparts (Table 2). Only steri-
cally modified derivatives 10 and 15 showed any differ-
ence when compared to their unsubstituted analogues,
delivering lower TON, presumably as a result of faster
initiation rates. Complex 16 displayed no activity what-
soever.
The electronically modified analogues delivered sim-
ilar TONs to each other. In the case of the sterically
modified (pre)catalysts, those that display higher initia-
tion rates, lead to a higher concentration of the 14-elec-
tron active species at any point in time and therefore ex-
hibit a faster decomposition rate resulting in lower TON.
In summary, electronic modification of the isopropox-
ystyrene led to (pre)catalysts that display different TOF
and deliver similar TONs when compared to each other,
Figure 3. TON for RCM of 5 with (pre)catalysts 1, 2, 3 and 4.
by different catalyst efficiencies. Studies by Grubbs and (pre)catalysts with steric modification of the isopro-
et al. have shown that where the concentration of the poxybenzylidene displaygreatly increased TOFand also
propagating methylidene is high, decomposition of the show a decrease in TON, delivering values similar to
complex is also increased.[12] In especially efficient cata- their parent phosphine-containing systems. Therefore
lysts (those with high initiation rates), the concentration the general trend of activity is 3¼12, 13, 14>4¼7, 8,
of the active 14-electron species at any point is high, and 9>1 ꢂ 15>2 ꢂ 10.
this could be connected to faster decomposition rates,
when compared to slower initiating species.[13]
Considering that the electronic effect of the chelating
isopropoxybenzylidene had no great effect on the TON,
With this in mind we examined the activities of steri- we were interested in investigating whether the TON
cally and electronically modified analogues of 3 and 4. could be improved in the case of the 1st generation cata-
We proposed that the initiation of the modified ana- lysts by exchange of the phosphine ligand. Two new
logues operate by the same mechanism. We expected (pre)catalysts were prepared (Figure 3) bearing differ-
(pre)catalysts 7 and 12, that carry an electron-donating ent phosphine ligands. The phosphines chosen were
substituent and therefore have lower efficiency than un- P(i-Pr)3, a sterically less demanding phosphine com-
substituted analogues 3 and 4, would display a higher ac- pared to P(Cy)3, but electronically similar, and the
tivity than the unsubstituted variants.
more sterically demanding cyclohexyl phoban ligand.[14]
Again RCM of 5 was used to test the hypothesis and Both catalysts were prepared in good yields using stand-
the reactions were carried out in an identical manner ard reaction conditions (see Experimental Section for
as tothose previouslyperformed. To guarantee reprodu- details), and their activity tested in the RCM reaction
cibility the results were each repeated four times (Ta- of 5.
ble 1).
From the results (Table 3) it can be seen that no signif-
Surprisingly, electronically modified (pre)catalysts 7– icant differences in activity between the (pre)catalyst 18
9 and 12–14 all delivered approximately the same TON and (pre)catalyst 3 were observed. These results suggest
Table 1. Parent ruthenium (pre)catalysts 3 and 4 and their sterically and electronically modified analogues.
L¼IHMES
Entry R1
L¼PCy3
R2
R3
R4
(Pre)catalyst
R1
R2
R3
R4
(Pre)catalyst
1
2
3
4
5
6
H
H
H
H
OMe
H
H
O-i-Pr
H
O-i-Pr O-i-Pr
H
H
H
H
O-i-Pr
H
H
H
H
H
4
7
8
9
10
H
H
H
H
OMe
H
H
O-i-Pr
H
O-i-Pr O-i-Pr
H
H
H
H
O-i-Pr
H
H
H
H
H
3
12
13
14
15
H
H
H
H
O-i-Pr 11[a]
O-i-Pr 16
[a]
Could not be synthesised.
Adv. Synth. Catal. 2005, 347, 1413–1422
asc.wiley-vch.de
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1415

Simon Maechling et al.
FULL PAPERS
Table 2. RCM of 5: TON of modified metathesis initiators.
[b]
Entry
(Pre)catalyst
Conversion [%][a]
TON
TOF [hꢀ1
]
1st
2nd
3rd
4th
Average
1
2
3
4
5
6
7
8
9
10
11
12
13
1
2
3
15
12
13
14
16
4
64
54
93
59
88
84
86
–
65
48
66
60
57
53
49
91
65
88
82
86
–
67
50
65
56
58
69
51
92
69
87
82
86
–
67
47
59
63
61
70
53
92
66
85
82
86
–
62
51
58
60
55
64
52
92
65
87
83
86
–
65
49
62
60
58
3200
2600
4600
3250
4350
4150
4300
–
3250
2450
3100
3000
2900
3000
130
60
214
34
103
n/d
–
60
150
11
48
n/d
10
7
8
9
[a]
Determined by HPLC.
See ref.[10]
[b]
a slightly lower activity than 3, delivering lower TONs
presumably due to its higher initiation rate arising
from the more sterically demanding cyclohexyl phoban
ligand.[17]
Following these results we investigated different sub-
strates for RCM in order to determine if the trends in
catalyst activity were general. The electronically modi-
fied (pre)catalysts had been shown not to display in-
creased TONs and the sterically modified catalysts dis-
played decreased TONs, so only (pre)catalysts 1, 2, 3
and 4 were investigated.
The next substrate investigated was 17 which yields an
8-membered system 18, previously used to determine
relative catalyst efficiencies.[10] Again, reactions were
carried out utilising previously developed conditions.
Figure 4. (Pre)catalysts 17 and 18.
that the less sterically demanding P(i-Pr)3 ligand[15] does The first unexpected observation was that the 1st gener-
not lead to increase in decomposition of the catalyst as ation catalysts gave much high conversions (and there-
very similar TONs were obtained when compared to fore TON) than their 2nd generation counterparts (Fig-
(pre)catalyst 3.[16] However, (pre)catalyst 17 displayed ure 5).
Table 3. RCM of 5 with (pre)catalysts 1, 3, 17 and 18.
Entry
Substrate
Catalyst
Loading [mol %]
Conversion [%][a]
TON
1
2
3
4
5
6
7
8
9
5
5
5
5
5
5
5
5
5
5
5
5
1
3
17
18
1
0.006
0.006
0.006
0.006
0.01
0.01
0.01
0.01
0.02
0.02
0.02
0.02
21
59
40
62
35
76
65
80
67
92
88
90
3500
9894
6667
10333
3581
7598
6500
8000
3372
4595
4400
4500
3
17
18
1
3
17
18
10
11
12
[a]
Conversion determined by HPLC.
1416
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1413–1422

TON Study Utilising Ruthenium-Based Olefin Metathesis Catalysts
FULL PAPERS
eration (pre)catalysts outperform the 2nd generation cata-
lysts.
Two further tosylamide substrates were investigated,
19 and 21, both delivering a 7-ring product. Investiga-
tions were carried out under the same conditions, utilis-
ing 0.01 mol % of the corresponding catalyst. The re-
sults are listed in Table 4, for purposes of comparison
the results for substrates 5 and 17 have been included.
The first point to note is that different substrates all
deliver different TONs, even when the catalysts are
compared at the same loading. When comparing the or-
der of substrates delivering the highest TON the order
follows 21>17>5>19. The difference between the
substrate reactivity of 19 and 21 and between that of 5
and 17 can be explained by considering the intermediate
formed during the RCM reaction. In both substrates
that show lower reactivity to RCM, 5 and 19, an intramo-
lecular coordination can occur as has previously been
described.[10]
From these results a general sequence of relative cat-
alyst activities can be derived. In the case of the sub-
strates 5 and 19 the activity of the catalysts follows the
trend 3ꢁ4>2 ꢂ 1; and in the case of substrates 17
and 21 the trend of activity follows the order 3ꢁ1>
4>2. In the case of substrates 17 and 21 the activity of
Figure 5. RCM of 17: Conversion as a function of (pre)cata-
lyst loading.
Interestingly, with (pre)catalysts 1 and 3 the conver- the 1st generation Grubbs catalyst 1 is higher than that
sion decreases by only approximately 10% when reduc-
ing the catalyst loadings from 0.1 to 0.01 mol %; howev-
er, precatalysts 2 and 4 show conversion losses of more
Table 4. TON of (pre)catalyst 1, 2, 3 and 4 in the RCM of to-
than 60% over the same region. Below catalyst loadings sylamides 5, 17, 19 and 21.
of 0.025 mol % the superiority of (pre)catalyst 3 be-
comes obvious. Utilising
a catalyst loading of
0.01 mol % delivers a conversion of 87%. To achieve
the same conversion with the 2nd generation catalysts re-
quires a loading of over ten times this amount.
Figure 6 displays the TON vs. S/C ratio. The catalysts
1, 2 and 4 all reach their maximum TON at around
0.01 mol % (S/C¼10000), however catalyst 3 reaches
its maximum TON (13000) at 0.006 mol %. With this
substrate there is a clear trend in activity of the catalysts
following the order 3>1>4>2. Importantly, the 1st gen-
Entry
Substrate
Catalyst
Conversion [%][a]
TON
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
5
5
5
5
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
37
36
76
49
46
12
87
24
4
3
16
5
60
51
80
54
3700
3600
7600
4900
4600
1200
8700
2400
400
300
1600
500
6000
5100
8000
5400
17
17
17
17
19
19
19
19
21
21
21
21
Figure 6. RCM of 17: TON compared to substrate catalyst ra-
tio.
[a]
Conversion determined by HPLC.
Adv. Synth. Catal. 2005, 347, 1413–1422
asc.wiley-vch.de
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1417

Simon Maechling et al.
FULL PAPERS
of the 2nd generation catalysts, however, with substrates RCM of substrates 5, 17, 19 and 21, demonstrates com-
5 and 19 it displays the lowest activity. The ruthenium parable activity to catalysts 2 and 4. Catalysts 2 and 4
complexes with isopropoxystyrene ligands afford higher also show comparable activity to each other, displaying
TONs than their phosphine-containing counterparts similar TONs as were observed with the tosylamides,
and most importantly (pre)catalyst 3 in all cases yields suggesting that the ester group has little effect on the ac-
the highest TON of all the (pre)catalysts.
tivity of these catalysts. It is interesting to note that
We were also interested in testing more challenging (pre)catalyst 4 performed slightly better than the other
substrates. The utilisation of RCM for the synthesis of catalysts in these examples and again the (pre)catalysts
cyclic amino acid derivatives was first reported by with a chelating isopropoxystyrene outperform their
Grubbs et al.[18] and we were interested in utilising these phosphine-containing counterparts in all cases.
more challenging substrates to examine the relative ac-
In the light of these results, we were interested to in-
tivities of the catalysts under investigation. As test sub- vestigate why substitution of a PCy3 ligand for a chelat-
strates we chose the allylglycine derivatives 22, 24 and 26 ing isopropoxystyrene leads to catalysts displaying high-
to furnish 6-, 7- and 8-membered ring systems. The com- er activity (2 and 4).
pounds display a structural resemblance to the N-tosyl-
In order to measure the relative catalyst stability a
amides, but contain an additional ester functionality study concerning the effect of temperature was under-
which has frequently been reported to coordinate and taken. It was hoped that these investigations would de-
deactivate the metal centre during the metathesis reac- liver guidelines for the relative thermal stability of the
tion.[19] Deactivation of the metal centre can occur precatalysts in solution in the presence of an RCM sub-
through coordination of the ester carbonyl to the metal strate. Utilisation of 0.01 mol % of catalyst with sub-
centre in the alkylidene intermediate.[20] The desired strate 5 and conducting the reactions in refluxing 1,2-di-
compounds were readily available through a three- chloroethane was investigated (Table 6). A sharp drop
step synthetic sequence, starting form commercially in TON was observed for 1 with conversion dropping
available allylglycine.[21]
from 66% to <25%. The same trend was found, but to
The RCM of the allylglycine derivatives were carried a lesser extent, for (pre)catalysts 2 and 3. However at
out under the standard conditions and the results have this elevated temperature (pre)catalyst 4 showed an
been summarized in Table 5.
identical result to the reaction in refluxing DCM, thus
Interestingly the lowest TONs were obtained with cat- demonstrating the higher thermal stability of this com-
alyst 1, suggesting poor tolerance to the ester functional- plex and (pre)catalyst 2 in solution when compared to
ity and a deactivation of the catalyst. (Pre)catalyst 3, the other (pre)catalysts. These results suggest that the
which previously displayed the highest activity in the stability of the formed active species of the catalyst is
not the only factor that effects the relative activities of
the (pre)catalysts.
Table 5. RCM of allylglycine derivatives 22, 24 und 26.
It has been suggested that the different activities of the
(pre)catalysts 3 and 4 carrying the isopropoxystyrene li-
gand, compared to their phosphine containing counter-
parts 1 and 2,[22] are due to the return of the disassociat-
ing ligand during the catalytic cyclic, e.g., PCy3 in the
Table 6. RCM of 5: Influence of the reaction temperature on
the activity of the (pre)catalysts 1, 2, 3 and 4.
Entry
Substrate
Catalyst
Conversion [%][a]
TON
1
2
3
4
5
6
7
8
9
22
22
22
22
24
24
24
24
26
26
26
26
1
2
3
4
1
2
3
4
1
2
3
4
11
17
15
16
33
42
36
48
36
49
46
54
1100
1700
1500
1600
3300
4200
3600
4800
3600
4900
4600
5400
Entry
Catalyst
TON[a]
TON[b]
1
2
3
4
1
3200
10000
2600
1200
6500
2300
3450
2[c]
3
10
11
12
4
3250
[a]
Determined after 14 hours in dichloromethane at 508C.
Determined after 14 hours in 1,2-dichloroethane at 808C.
0.006 mol % (pre)catalyst utilised.
[b]
[c]
[a]
Determined by HPLC.
1418
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1413–1422

TON Study Utilising Ruthenium-Based Olefin Metathesis Catalysts
FULL PAPERS
case of 1 and 2 and the isopropoxybenzylidene in the
From the results obtained it can be suggested that, un-
case of 3 and 4. It was assumed that following its dissoci- der the reaction conditions used, reassociation of nei-
ation the free phosphine can then re-coordinate to the ther the PCy3 ligand nor the isopropoxybenzylidene oc-
metal centre in preference to an olefin binding, and curs and that these pathways have no influence on the
hence decrease activity. We recognise that the existence activity of the precatalysts. Assuming that the propagat-
of free phosphine in solution under the reaction condi- ing species emerging from 1 and 3, 2 and 4 are identical,
tions could possibly account for the lower relative activ- what could the reasons be for their displayed differences
ity of 1 and 2 when compared to 3 and 4. By carrying out in activity? One possible explanation could lie in differ-
the RCM of substrate 5 in the presence of excess CuCl, a ent initiation mechanisms between 1 and 3, 2 and 4.
phosphine scavenger, phosphine reassociation would be Scheme 1 represents a mechanism for these events.
prevented, allowing the rebinding of phosphine during
Firstly the activation of the precatalysts 3 and 4, as
the reaction to be investigated through the TON ob- postulated by Hoveyda et al.,[22] proceeds by opening
ꢀ
tained in the presence and absence of free phosphine. of the Ru O chelate and the development of a vacant
When reactions were run under the standard conditions coordination site and intermediates 3a and 4a. Concom-
in the presence of CuCl, the TON only differed minimal- itantly a rotation takes place around the C1 C2 bond
ly when compared to those experiments undertaken and the isopropoxy group rotates into an orientation
without CuCl present in the reaction, suggesting that un- away from the metal centre. In the case of 3a, we postu-
der the conditions used to determine the TON the cata- late that this forms a discrete intermediate. This sugges-
lysts 1and 2are not influenced by the reassociation of the tion is supported by complex 16, which is readily syn-
ꢀ
phosphine ligand during the catalytic cycle.
thesised from 1, but displays no metathesis reactivity,
It has also been proposed that reassociation of the iso- presumably because upon rotation of the chelating iso-
proxystyrene ligand during the catalytic cycle, regener- propoxystyrene no open coordination site is available.
ating the starting precatalyst, could be the reason for However, the analogous 2nd generation complex 11
the increased activity of 3 and 4 when compared to 1 could not be isolated, presumably due to unfavourable
and 2. To test this theory the RCM of substrate 5 with steric interactions between the isopropoxy ether and
1 and 2 was undertaken in the presence of an equimolar the sterically demanding IHMES (Figure 7).
quantity of the isopropoxystyrene in the presence of
The formed species 3a and 4a are in equilibrium with
CuCl (Table 7). It was thought that following dissocia- stable precatalysts 3 and 4. In contrast to the precatalysts
tion of the PCy3, which would be captured by the excess 3 and 4, activation of 1 and 2 proceeds directly following
CuCl in the reaction mixture, the isopropoxystyrene li- dissociation of the PCy3 ligand to give intermediate 1a or
gand could coordinate and generate in situ (pre)cata- 2a,[23] with decomposition of this intermediate in compe-
lysts 3 and 4, and therefore afford increased TON. Inter- tition with olefin binding. The differences in the activi-
estingly, the same TONs were obtained as was seen ties between both precatalysts with a chelating isopro-
when no additives were used. From these results it can poxystyrene and the analogous PCy3-containing cata-
be suggested that under the reaction conditions utilised lysts and between the 1st and 2nd generation (pre)cata-
reassociation of the chelating isopropoxystyrene has no lysts can be explained by the different lifetimes and sta-
decisive influence on the activities of the precatalysts 3 bilities of the emerging intermediates, and the resulting
and 4.
concentrations of the 14-electron propagating species 29
or 30.
While the alkylidene complexes 3a and 4a are in equi-
librium with the stable precatalysts 3 and 4, the coordi-
nation of the olefins to the complex is in competition
with the decomposition of the 14-electron species 1a
and 2a. We believe that 1 and 2 provide a higher concen-
Table 7. RCM of 5: Influence of the reaction conditions on
the activity of the (pre)catalysts 1, 2, 3 and 4.
Entry
Catalyst
TON
TON[a]
TON[a],[b]
1
2
3
4
1
3300
10000
2800
3150
9667
2650
3350
3200
–
2600
–
2[c]
3
4
3300
[a]
Three equivalents CuCl added.
One equivalent ortho-isopropoxystyrene added.
0.006 mol % (pre)catalyst utilised.
[b]
[c]
Figure 7. Diagram to represent sterically demanding IHMES
compared to (pre)catalyst 16.
Adv. Synth. Catal. 2005, 347, 1413–1422
asc.wiley-vch.de
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1419

Simon Maechling et al.
FULL PAPERS
Scheme 1. Mechanism for RCM of terminal alkenes.
tration of the propagating methylidene species and ac- vided unexpected results. In some examples presented
cordingly display a lower activity when compared to pre- the 1st generation catalysts gave higher TONs than the
catalysts 3 and 4. The more sterically demanding so-called more active 2nd generation catalysts although
IHMES ligand in comparison to PCy3,[24] could also im- displaying very different initiation profiles, and consis-
pede the coordination of olefins to intermediate 4a, tently afford higher results with respect to TON.
leading to higher decomposition rates of this ruthenium
complex in comparison to 3a.
The order of activity of the (pre)catalysts can be ra-
tionalised through the stability and the delivered con-
Therefore (pre)catalysts 1 and 2 produce a higher con- centration of the propagating species. The stability and
centration of the propagating methylidene, particularly activity of the intermediate reactive species is substan-
when compared to (pre)catalyst 3. The reason behind tially altered by the steric and electronic characteristics
the increased activity of 4 compared to 2 can be under- of the neutral ligand remaining at the metal centre. In all
stood by the decomposition pathway recently pro- examples (pre)catalysts 3 and 4 displayed higher activity
posed.[1] With no phosphine present, (pre)catalyst 3 than their phosphine-containing counterparts 1 and 2.
can display a longer propagating lifetime. (Pre)catalyst Further investigations including the synthesis of less
3 can be regarded as a ꢀslow releaseꢁ system, where the sterically demanding NHC ligands are currently under-
metal is delivered to the olefin, affording only small way.
amounts of the propagating species at any one time
and therefore minimising bimolecular decomposition
pathways, therefore maximising TON.
Experimental Section
Conclusion
General Remarks
A study of TON in solution utilising ruthenium-based
All catalysts were prepared according to the previously pub-
olefin metathesis catalysts has been completed and pro- lished standard procedure.[10]
1420
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1413–1422

TON Study Utilising Ruthenium-Based Olefin Metathesis Catalysts
FULL PAPERS
Synthesis of 17
Acknowledgements
The general procedure was followed using [(PhobCy)2(Cl)2-
S. M. wishes to thank the Graduiertenkolleg “Synthetische,
mechanistische und reaktionstechnische Aspekte von Metallka-
talysatoren” for financial support. We also would like to thank
the ꢀFonds der Chemischen Industrieꢁ for financial support.
[17]
¼
Ru CHPh]
(87 mg, 0.123 mmol), 2-isopropoxystyrene
(40 mg, 0.246 mmol) and CuCl (12 mg, 0.121 mmol) in CH2Cl2
(7 mL). Column chromatography (CH2Cl2) gave 17 as a brown
1
solid; yield: 45 mg (68%). H NMR (CDCl3, 250 MHz): d¼
17.64 (d, J¼1.5 Hz, 1H), 7.66 (m, 2H), 7.08 (m, 2H), 5.30 (m,
1H), 2.79 (m, 2H), 2.49–1.31 (series of multiplets, 29H); 13C
NMR (CDCl3, 125 MHz): d¼283.3 (CH), 152.7 (C), 144.7
(C), 129.9 (CHAr), 123.0 (CHAr), 122.8 (CHAr), 113.4
(CHAr), 75.6 (CH), 33.4 (CH), 29.0 (CH2), 28.3 (CH2), 27.7–
20.9 (phoban ligand resonances); 31P (CDCl3, 200 MHz): d¼
37.64; IR (Nujol): n¼2929, 2855, 1702, 1589, 1451 cmꢀ1; HR-
MS: calcd: 544.0996; found: 544.0991.
References
[1] a) G. C. Bazan„ J. H. Oskam, H. N. Cho, L. Y. Park,
R. R. Schrock, J. Am. Chem. Soc. 1991, 113, 6889–
6907; b) R. R. Schrock, Tetrahedron 1999, 55, 8141–
8153; c) R. R. Schrock, Acc. Chem. Res. 1990, 23, 158–
165; d) R. R. Schrock, A. H. Hoveyda, Angew. Chem.
Int. Ed. 2003, 42, 4592–4633.
Synthesis of 18
[2] T. M. Trnka, R. H. Grubbs, Acc. Chem. Res. 2001, 34,
18–29.
[3] M. S. Sanford, M. Ulman, R. H. Grubbs, J. Am. Chem.
Soc. 2001, 123, 749–750.
[4] J. S. Kingsbury, J. P. A. Harrity, A. H. Hoveyda, J. Am.
Chem. Soc. 1999, 121, 791–799.
[5] a) S. Gessler, S. Randl, S. Blechert, Tetrahedron Lett.
2000, 41, 9973–9976; b) S. B. Garber, J. S. Kingsbury,
B. L. Gray, A. H. Hoveyda, J. Am. Chem. Soc. 2000,
122, 8168–8179.
[6] a) K. Grela, S. Harutyunyan, A. Michrowska, Angew.
Chem. Int. Ed. 2002, 41, 4038–4040; b) J. O. Krause, O.
Nuyken, K. Wurst, M. R. Buchmeiser, Chem. Eur. J.
2004, 10, 777–784; c) M. A. O. Volland, S. M. Hansen,
F. Rominger, P. Hofmann, Organometallics 2004, 23,
800–816; d) J. A. Love, J. P. Morgan, T. M. Trnka, R. H.
Grubbs, Angew. Chem. Int. Ed. 2002, 41, 4035–4037;
e) J. Louie, R. H. Grubbs, Angew. Chem. Int. Ed. 2001,
40, 247–249.
[7] a) J. A. Love, J. P. Morgan, T. M. Trnka, R. H. Grubbs,
Angew. Chem. Int. Ed. 2002, 41, 4207–4209; b) A. Furst-
ner, L. Ackermann, B. Gabor, R. Goddard, C. W. Leh-
mann, R. Mynott, F. Steltzer, O. R. Thiel, Chem. Eur. J.
2001, 7, 3236–3253; c) R. Bujok, M. Bieniek, M. Masnyk,
A. Michrowska, A. Sarosiek, H. Stepowska, D. Arlt, K.
Grela, J. Org. Chem. 2004, 69, 6894–6896; d) P. E. Ro-
mero, W. E. Piers, R. McDonald, Angew. Chem. Int.
Ed. 2004, 43, 6161–6165; e) H. Wakamatsu, S. Blechert,
Angew. Chem. Int. Ed. 2002, 41, 2403–2405.
[8] M. B. Dinger, J. C. Mol, Adv. Synth. Catal. 2002, 344,
671–677.
The general procedure was followed using [(P-i-Pr3)2(Cl)2-
[25]
¼
Ru CHPh]
(37 mg, 0.063 mmol), 2-isopropoxystyrene
(20 mg, 0.127 mmol) and CuCl (6 mg, 0.063 mmol) in CH2Cl2
(4 mL). Column chromatography (CH2Cl2) gave 18 as a brown
1
solid; yield: 15 mg (50%). H NMR (CDCl3, 500 MHz): d¼
17.39 (d, J¼4.5 Hz, 1H), 7.68–7.59 (m, 2H), 7.11–7.05 (m,
2H), 5.28 (pentet, J¼6.5 Hz, 1H), 2.70–2.59 (m, 3H), 1.81 (d,
J¼6.5 Hz, 6H) 1.46, (q, J¼7.2 Hz, 3H); 13C NMR (CDCl3,
125 MHz): d¼279.6 (CH), 153.0 (C), 144.0 (C), 129.7
(CHAr), 122.7 (CHAr), 122.6 (CHAr), 113.5 (CHAr), 25.7
(CHP), 20.1 (CH3); 31P NMR (CDCl3, 200 MHz): d¼57.51;
IR (Nujol): n¼2963, 2932, 2874, 1588, 1475, 1452 cmꢀ1; HR-
MS: calcd: 480.0683; found: 480.0690.
General Procedure for RCM
All substrates were synthesised according to previously pub-
lished methods,[26] and purified by column chromatography
(twice) before kugelrohr distillation. All substrates were fresh-
ly distilled immediately before use.
An accurately weighed sample (six-figure analytical bal-
ance) of solid catalyst was dissolved in toluene to obtain a con-
centration of 1 mg/1 mL. Well determined volumes of these
solutions were introduced via a Gilson Pipetteman under nitro-
gen to the preheated (508C) 0.05 M solutions of substrate in
DCM under an inert atmosphere (N2). The reaction mixtures
were stirred at reflux for 14 hours and the progress of the meta-
thesis reactions was measured by HPLC.
HPLC: Waters 600; Autosampler Waters 717; column: RP-7
C-18, (4 mm), particle size: 7 mm; PDA detector Waters 991,
l¼254 nm. MeOH/H2O¼8:2; flow rate 0.8 mL/min; reten-
tion times (min): 3.49 (6), 4.59 (5), 5.88 (toluene). MeOH/H2
O¼8:2; flow rate 0.8 mL/min; retention times (min): 4.19
(18), 4.80 (toluene), 5.41 (17). MeOH/H2O¼7:3; flow rate
1.5 mL/min; retention times (min): 3.52 (20), 4.87 (toluene),
6.23 (19). MeOH/H2O¼7:3; flow rate 1.5 mL/min; retention
times (min): 3.62 (22), 6.10 (21), 7.42 (toluene). MeOH/H2
O¼8:2; flow rate 1.2 mL/min; retention times (min): 3.30
(23), 4.39 (22), 5.56 (toluene). MeOH/H2O¼9:1; flow rate
0.8 ml/min; retention times (min): 4.25 (25), 4.94 (24), 5.81 (tol-
uene). MeOH/H2O¼9:1; flow rate 0.8 mL/min; retention
times (min): 4.58 (27), 5.14 (26), 5.73 (toluene).
[9] J. A. Love, M. S. Sanford, M. W. Day, R. H. Grubbs, J.
Am. Chem. Soc. 2003, 125, 10103–10109.
[10] M. Zaja, S. Connon, A. M. Dunne, M. Rivard, N. Busch-
mann, J. Jiricek, S. Blechert, Tetrahedron 2003, 59, 6545–
6558.
[11] M. S. Sanford, J. A. Love, in: Handbook of Metathesis,
(Ed.: R. H. Grubbs), Wiley-VCH, Weinheim, 2003; Vol.
1, pp 112–131.
[12] M. Ulman, R. H. Grubbs, J. Org. Chem. 1999, 64, 7202–
7207.
[13] S. Connon, M. Rivard, M. Zaja, S. Blechert, Adv. Synth.
Catal. 2003, 345, 572–575.
Adv. Synth. Catal. 2005, 347, 1413–1422
asc.wiley-vch.de
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1421

Simon Maechling et al.
FULL PAPERS
[14] H. C. L. Abbenhuis, U. Burckhardt, V. Gramlich, C.
Koellner, P. S. Pregosin, R. Salzmann, S. Togni, Organo-
metallics 1995, 14, 759–766.
[15] C. A. Tolman, Chem. Rev. 1977, 77, 313.
[16] M. Ulmann, R. H. Grubbs, J. Org. Chem. 1999, 64, 7202–
7207.
[17] G. S. Forman, A. E. McConnell, M. J. Hanton, A. M. Z.
Slawin, R. P. Tooze, W. J. van Rensburg, W. H. Meyer,
C. Dwyer, M. M. Kirk, D. W. Serfontein, Organometal-
lics 2004, 23, 4824–4827.
[21] S. Varray, C. Gauzy, F. Lamaty, R. Lazaro, J. Martinez, J.
Org. Chem. 2000, 65, 6787–6790.
[22] A. H. Hoveyda, D. G. Gillingham, J. J. Van Veldhuizen,
O. Kataoka, S. B. Garber, J. S. Kingsbury, J. P. A. Harrity,
Org. Biomol. Chem. 2004, 2, 1–16.
[23] a) M. S. Sanford, M. Ulman, R. H. Grubbs, J. Am. Chem.
Soc. 2001, 123, 749–750; b) E. L. Dias, S. T. Nguyen,
R. H. Grubbs, J. Am. Chem. Soc. 1997, 119, 3887–3897;
c) M. S. Sanford, J. A. Love, R. H. Grubbs, J. Am.
Chem. Soc. 2001, 123, 6543–6554.
[18] a) S. J. Miller, R. H. Grubbs, J. Am. Chem. Soc. 1995,
117, 5855–5856; b) S. J. Miller, H. E. Blackwell, R. H.
Grubbs, J. Am. Chem. Soc. 1996, 118, 9606–9614.
[19] a) A. Fürstner, Top. Organomet. Chem. 1998, 1, 37–72;
b) A. Fürstner, K. Langemann, Synthesis 1997, 792–803.
[20] A. Fürstner, O. R. Thiel, C. W. Lehmann, Organometal-
lics 2002, 21, 331–335.
[24] a) B. K. Campion, R. H. Heyn, T. D. Tilley, Chem. Com-
mun. 1988, 278–280; b) J. Huang, H. J. Schanz, E. D. Ste-
vens, S. P. Nolan, Organometallics 1999, 18, 2370–2375.
[25] E. L. Dias, S. T. Nguyen, R. H. Grubbs, J. Am. Chem.
Soc. 1997, 119, 3887–3897.
[26] S. Varray, C. Gauzy, F. Lamaty, R. Lazaro, J. Martinez, J.
Org. Chem. 2000, 65, 6787–6790.
1422
ꢂ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1413–1422