Novel ruthenium-based metathesis catalysts containing electron-withdrawing ligands: Synthesis, immobilization, and reactivity

Article abstract of DOI:10.1021/jo0477594

The syntheses and reactivity of seven different ruthenium-based metathesis catalysts are described. Ru(CF₃COO)₂(PCy ₃)(=CH-2-(2-PrO)C₆H₄) (1),Ru(CF ₃COO)₂(1,3-dimesityldihydroimi

Full text of DOI:10.1021/jo0477594

Novel Ruthenium-Based Metathesis Catalysts Containing Electron-
Withdrawing Ligands: Synthesis, Immobilization, and Reactivity
Tobias S. Halbach,† Stefan Mix,‡ Dirk Fischer,‡ Simon Maechling,‡ Jens O. Krause,†
Carsten Sievers,§ Siegfried Blechert,*,‡ Oskar Nuyken,*,† and Michael R. Buchmeiser*,|
Lehrstuhl fu¨r Makromolekulare Stoffe, Technische Universita¨t Mu¨nchen, Lichtenbergstr. 4,
D-85747 Garching, Germany, Institut fu¨r Organische Chemie, Technische Universita¨t Berlin,
Strasse des 17. Juni 135, D-10623 Berlin, Germany, Lehrstuhl II fu¨r Technische Chemie,
Technische Universita¨t Mu¨nchen, Lichtenbergstr. 4, D-85747 Garching, Germany, and
Leibniz Institut fu¨r Oberfla¨chenmodifizierung (IOM) e. V., and Institut fu¨r Technische Chemie,
Universita¨t Leipzig, Permoserstr. 15, D-04318 Leipzig, Germany
oskar.nuyken@ch.tum.de; blechert@chem.tu-berlin.de; michael.buchmeiser@iom-leipzig.de
Received December 22, 2004
The syntheses and reactivity of seven different ruthenium-based metathesis catalysts are described.
Ru(CF3COO)2(PCy3)(dCH-2-(2-PrO)C6H4) (1), Ru(CF3COO)2(1,3-dimesityldihydroimidazolin-2-ylidene)-
(dCH-2-(2-PrO)C6H4) (2), and Ru(CF3COO)2(PCy3)(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-
C6H5) (3) were prepared via chlorine exchange by reacting RuCl2(PCy3)2(dCH-2-(2-PrO)C6H4), RuCl2-
(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)C6H4), and RuCl2(PCy3)(1,3-dimesityldihy-
droimidazolin-2-ylidene)(dCHC6H5), respectively, with silver trifluoroacetate (Cy ) cyclohexyl). In
analogy, Ru(CF3CF2COO)2(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)C6H4) (4) and Ru-
(CF3CF2CF2COO)2(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)C6H4) (5) were prepared
from RuCl2(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)C6H4) via reaction with CF3CF2-
COOAg and CF3CF2CF2COOAg, respectively. Ru(C6F5COO)2(1,3-dimesityldihydroimidazolin-2-ylidene)-
(dCH-2-(2-PrO)C6H4) (6) and Ru(C6F5O)2(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)-
C6H4) (7) were prepared from RuCl2(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)C6H4)
via reaction with C6F5COOTl and C6F5OTl, respectively. Supported catalysts Ru(PS-DVB-CH2OOCCF2-
CF2CF2COO)(CF3COO)(PCy3)(1,3-dimesityldihydroimidazolin-2-ylidene)(dCHC6H5) (8), Ru(PS-DVB-
CH2OOCCF2CF2CF2COO)(CF3COO)(PCy3)(dCH-2-(2-PrO)C6H4) (9), and Ru(PS-DVB-CH2OOCCF2-
CF2CF2COO)(CF3COO)(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)C6H4) (10) were
synthesized by reaction of RuCl2(PCy3)(1,3-dimesityldihydroimidazolin-2-ylidene)(dCHC6H5), RuCl2-
(PCy3)(dCH-2-(2-PrO)C6H4), and RuCl2(1,3-dimesityldihydroimidazolin-2-ylidene)(dCH-2-(2-PrO)C6H4),
respectively, with a perfluoroglutaric acid-derivatized poly(styrene-co-divinylbenzene) (PS-DVB) support
(silver form). Halogen exchange in PCy3-containing systems had to be carried out in dichloromethane
in order to suppress precipitation of AgCl‚PCy3. The reactivity of all new catalysts in ring-closing
metathesis (RCM) of hindered electron-rich and -poor substrates, respectively, at elevated temperature
(45 °C) was compared with that of existing systems. Diethyl diallylmalonate (DEDAM, 11), diethyl
allyl(2-methylallyl)malonate (12), N,N-diallyl-p-toluenesulfonamide (13), N-benzyl-N-but-1-en-4-ylbut-
2-enecarboxylic amide (14), and N-allyl-N-(1-carboxymethyl)but-3-en-1-yl-p-toluenesulfonamide (15)
were used as educts. Supported catalysts were prepared with high loadings (2.4, 22.1, and 160 mg of
catalyst/g PS-DVB for 8, 9, and 10, respectively). Catalyst 8 showed higher and catalysts 9 and 10
sowed significantly reduced activities in RCM compared to their homogeneous analogues. Thus, with
8, turnover numbers (TONs) up to 4200 were realized in stirred-batch (carousel) RCM experiments.
To elucidate the nature of the bound species, catalysts 8-10 were subjected to 13C- and 31P-MAS NMR
spectroscopy. These investigations provided evidence for the proposed structures. Leaching of ruthenium
into the reaction mixture was low, resulting in ruthenium contents <85 ppb (ng/g) in the final RCM-
derived products.
Introduction
ter.1 Thus, replacing phosphines by highly basic N-
heterocyclic carbenes (NHCs) increased reactivity by
many orders.2,3 Similarly, the introduction of the “Hov-
eyda-ligand”, i.e., the [dCH-2-(2-PrOC6H4)] carbene ligand
resulted in metathesis catalysts with increased stability
and excellent recyclability.4,5 Further variations and
The development of highly active ruthenium-based
catalysts for various applications in metathesis-derived
reactions proceeded in an evolutionary way by optimizing
the ligand sphere around the corresponding metal cen-
10.1021/jo0477594 CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/05/2005
J. Org. Chem. 2005, 70, 4687-4694
4687

Halbach et al.
improvements of this ligand were reported by the Blech-
ert and Grela groups.6-15 In a joint research project,
Nuyken and Buchmeiser et al. recently reported on the
synthesis of various ruthenium-based metathesis cata-
lysts prepared by the replacement of both chlorine atoms
in RuCl2(NHC)(CHR) (NHC ) 1,3-dimesitylimidazol-2-
ylidene, 1,3-dimesityl-4,5-dihydroimidazolin-2-ylidene,
1,3-dimesityl-3, 4, 5, 6-tetrahydropyrimidin-2-ylidene, R
) 2-(2-PrO)-C6H4, 2-(2-PrO)-5-NO2C6H3) by carboxy-
lates.16-20 Where fluorinated carboxylates were used, a
dramatic increase in reactivity was observed, even al-
lowing the use of these systems in the cyclopolymeriza-
tion of 1,6-heptadiynes.18,21 On the basis of the reaction
of polymer-bound silver carboxylates, the immobilization
of various catalysts was accomplished. However, chem-
istry was so far restricted to phosphine-free systems,
since phosphine-containing catalysts suffered from lack
of stability due to the formation of AgCl‚PCy3 during
synthesis. Our aim was to apply the concept of chlorine
replacement by ligands containing electron-withdrawing
groups to phosphine catalysts. In this paper, we report
on the synthesis of metathesis catalysts based on PCy3
and IMesH2 ligands (PCy3 ) tricyclohexylphosphine,
IMesH2 ) 1,3-dimesityldihydroimidazolin-2-ylidene) ac-
cessible via replacement of both chlorines by trifluoro-
acetate groups. In addition, heterogenization on poly-
(styrene-co-divinylbenzene) resins is described. The nature
of the polymer-bound species was elucidated by 1H-, 13C-
and 31P-MAS NMR spectroscopy. To benchmark the new
systems in terms of their reactivity, they were all
subjected to various RCM reactions using diethyl dial-
lylmalonate (DEDAM, 11), diethyl allyl(2-methylallyl)-
malonate (12), N,N-diallyl-p-toluenesulfonamide (13),
N-benzyl-N-but-1-en-4-ylbut-2-enecarboxylic amide (14),
and N-allyl-N-(1-carboxymethyl)but-3-en-1-yl-p-toluene-
sulfonamide (15) as substrates. In this paper, the syn-
thesis of the new catalysts, their structure, and catalytic
activity shall be outlined in detail. Catalyst loadings were
chosen in a way that both the maximum TON as well as
the amount of catalyst necessary for a conversion >90%
were determined.
FIGURE 1. Structures of the “Grubbs-Hoveyda 1st genera-
tion”, “Grubbs-Herrmann” (“Grubbs’ 2nd generation”), and
“Grubbs-Hoveyda 2nd generation” catalyst (IMesH2 ) 1,3-
dimesityldihydroimidazolin-2-ylidene).
established proocedure with phosphine-free catalysts,19,20
substitution of chlorines in phosphine-based systems has
not been realized yet. This is a direct consequence of the
formation of AgCl‚PCy3 in the course of the reaction,
which was carried out in THF so far in order to dissolve
the silver carboxylates.19,20 However, in the case where
the (phosphane-containing) catalyst precursor was added
to a solution of the corresponding silver carboxylate in
methylene chloride instead of THF, a clean reaction
occurred.
Apparently, the noncoordinating solvent methylene
chloride dissolved only small amounts of the silver salt,
thus preventing the formation of AgCl‚PCy3. Keeping this
particularity in mind, catalysts 1-5 were prepared in
yields up to 98% (Scheme 1). To investigate the influence
of even more electron-withdrawing and sterically de-
manding ligands, catalysts 6 and 7 were synthesized. For
their synthesis, the corresponding thallium salts had to
be used. Apart from toxicity, some advantages have to
be mentioned. On one hand, thallium salts are stable
toward light; on the other hand, synthesis is simplified,
(1) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29.
(2) Scholl, M.; Trnka, T. M.; Morgan, J. P.; Grubbs, R. H. Tetrahe-
dron Lett. 1999, 40, 2247-2250.
(3) Bielawski, C. W.; Grubbs, R. H. Angew. Chem. 2000, 112, 3025-
3028.
(4) Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P. J., Jr.;
Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 791-799.
(5) Gessler, S.; Randl, S.; Blechert, S. Tetrahedron Lett. 2000, 41,
9973-9976.
(6) Randl, S.; Buschmann, N.; Connon, S. J.; Blechert, S. Synlett
2001, 1547-1450.
(7) Wakamatsu, H.; Blechert, S. Angew. Chem. 2002, 114, 832-834.
(8) Connon, S. J.; Dunne, A. M.; Blechert, S. Angew. Chem. 2002,
114, 3989-3993.
Results and Discussion
(9) Wakamatsu, H.; Blechert, S. Angew. Chem. 2002, 114, 2509-
2511.
Synthesis and Structure of Ru Complexes 1-7.
The phosphine-containing compounds “Grubbs-Hoveyda
1st generation” and “Grubbs-Herrmann” (“Grubbs 2nd
generation”) catalyst as well as the phosphine-free
Grubbs-Hoveyda 2nd generation catalyst (Figure 1) were
used as precursors for the synthesis of the new chlorine-
free catalysts.
Though substitution of one chlorine by alkyl carboxy-
lates and substitution of both chlorines by trifluo-
romethanesulfonate or other electron-withdrawing ligands
such as fluoroalkyl carboxylates is meanwhile a well-
(10) Zaja, M.; Connon, S. J.; Dunne, A. M.; Rivard, M.; Buschmann,
N.; Jiricek, J.; Blechert, S. Tetrahedron 2003, 59, 6545-6558.
(11) Buschmann, N.; Wakamatsu, H.; Blechert, S. Synlett 2004,
667-670.
(12) Grela, K.; Harutyuyan, S.; Michrowska, A. Angew. Chem. 2002,
114, 4210-4212.
(13) Grela, K.; Kim, M. Eur. J. Org. Chem. 2003, 963-966.
(14) Michrowska, A.; Bujok, R.; Harutyunyan, S.; Sashuk, V.;
Dolgonos, G.; Grela, K. J. Am. Chem. Soc. 2004, 126, 9318-9325.
(15) Bujok, R.; Bieniek, M.; Masnyk, M.; Michrowska, A.; Sarosiek,
A.; Stepowska, H.; Arlt, D.; Grela, K. J. Org. Chem. 2004, 69, 6894-
6896.
(16) Krause, J. O.; Lubbad, S.; Mayr, M.; Nuyken, O.; Buchmeiser,
M. R. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 2003, 44,
790-791.
(17) Krause, J. O.; Lubbad, S.; Nuyken, O.; Buchmeiser, M. R. Adv.
Synth. Catal. 2003, 345, 996-1004.
* To whom correspondence should be addressed. (O.N.) Fax: +49-
(0)89-289-13571. (S.B.) Fax: +49-(0) 30-314-23619. (M.R.B.) Fax: +49-
(0)341-235-2584.
(18) Krause, J. O.; Zarka, M. T.; Anders, U.; Weberskirch, R.;
Nuyken, O.; Buchmeiser, M. R. Angew. Chem. 2003, 115, 6147-6151.
(19) Krause, J. O.; Lubbad, S. H.; Nuyken, O.; Buchmeiser, M. R.
Macromol. Rapid Commun. 2003, 24, 875-878.
(20) Krause, J. O.; Wurst, K.; Nuyken, O.; Buchmeiser, M. R. Chem.
Eur. J. 2004, 10, 778-785.
† Lehrstuhl fu¨r Makromolekulare Stoffe, Technische Universita¨t
Mu¨nchen.
‡ Institut fu¨r Organische Chemie, Technische Universita¨t Berlin.
§ Lehrstuhl II fu¨r Technische Chemie, Technische Universita¨t
Mu¨nchen.
| Leibniz Institut fu¨r Oberfla¨chenmodifizierung and Institut fu¨r
Technische Chemie, Universita¨t Leipzig.
(21) Krause, J. O.; Nuyken, O.; Buchmeiser, M. R. Chem. Eur. J.
2004, 10, 2029-2035.
4688 J. Org. Chem., Vol. 70, No. 12, 2005

Ruthenium-Based Metathesis Catalysts
SCHEME 1. Synthesis of Catalysts 1-7
TABLE 1. Results for the RCM of Diethyl
Diallylmalonate (DEDAM, 11) with Catalysts 1, 2, 4-7,
and 8-10a
as TlOEt deprotonates and forms the thallium salt in
situ. Using this approach, compounds 6 and 7 were
prepared in 87 and 85% yield, respectively. Generally,
catalysts 1-7 are stable in the solid state even upon
exposure to air; however, solutions (particularly diluted
ones) cannot be stored for an unlimited amount of
time.
catalyst
mol % of compd
yield (%)
TON
Grubbs 1st
0.02
0.02
0.1
0.01
0.02
0.02
0.02
0.1
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.02
0.02
0.1
53
81
>99
61
85
91
<1
4
43
9
50
67
51
8
37
54
<1
4
6
18
2650
4050
1000
6100
4250
4550
Grubbs-Hoveyda 1st
Grubbs-Hoveyda 1st
Grubbs 2nd
Grubbs 2nd
Reactivity of Catalysts 1-7 in Ring-Closing
Metathesis (RCM). Turnover numbers (TONs) rather
than yields are given as a measure for reactivity. Though
organic chemists are more interested in yields when
running a particular reaction, TONs allow for estimating
the reactivity of a catalyst, particularly in the case where
yields are <85%. In this case, TONs represent the
maximum reactivity attainable in the chosen solvent at
a certain concentration, catalyst-to-substrate ratio, and
temperature and are therefore the preferred indicator for
catalyst evaluation. In a first set of reactions, catalysts
1-7 were used for the RCM of diethyl diallylmalonate
(DEDAM, 11). As can be deduced from Table 1, the
phosphane-based catalyst 1 showed basically no reactiv-
ity at all. Reactivity was also comparably low for 3 (Table
1); therefore, its use was limited to the reaction with
DEDAM. Good reactivity in terms of turnover numbers
(TONs) was observed for the NHC-based catalysts 2 and
4-7. In particular, catalyst 5 showed high reactivity;
however, none of the systems could rival the reactivity
of the parent systems (Grubbs-Hoveyda 1st generation”,
“Grubbs 2nd generation”, Grubbs-Hoveyda 2nd genera-
tion catalyst). Similar results were obtained in the RCM
reaction of diethyl allyl-(2-methylallyl)malonate (12) and
N,N-diallyl-p-toluenesulfonamide (13), Scheme 2, Tables
2 and 3) Surprisingly, the reactivity of catalysts 1, 2, and
Grubbs-Hoveyda 2nd
1
1
40
2150
450
2500
3350
2550
400
2
3
4
5
6
7
8
3660
2680
8
9
9
40
300
180
10
10
0.02
0.1
a Conditions: solvent, CH2Cl2; T ) 45 °C; t ) 18 h.
4-6 for N-benzyl-N-but-1-en-4-ylbut-2-enecarboxylic amide
(14) was quite low. In contrast, 1, 2, and 6 showed good
reactivity in terms of TONs for N-allyl-N-(1-carboxy-
methyl)but-3-en-1-yl-p-toluenesulfonamide (15), with cata-
lyst 2 almost rivaling the parent catalysts. Again in terms
of TONs, the use of benzoate ligands, yielding 6, results
in a catalyst with good reactivity for DEDAM, 12, and
13 yet modest reactivity for 14 and 15. However, sub-
stitution of the chlorines with a perfluorated phenoxide
dramatically decreased reactivity. In fact, reactivity
J. Org. Chem, Vol. 70, No. 12, 2005 4689

Halbach et al.
SCHEME 2. Summary of RCM Experiments with
Monomers 11-15 Carried out with Catalysts 1-9
(C6F5O)(IMes)(CHPh) and TONs up to 40 000 in the RCM
of DEDAM were reported. Since no further RCM data
were reported for this catalyst, it is hard to judge whether
these differences are substrate specific or a result of the
combination of the pyridine with IMes ligand. So far,
effects of replacement of the chlorines by other ligands
on reactivity are still based on serendipity. Thus, Grubbs
et al. performed halogen variations with the RuCl2(d
CHdCHCPh2)(PCy3)2 system.23 Replacement of both
chlorine ligands by bromine, iodine, or trifluoroacetate
lead to less active or less stable catalysts. In contrast,
changing the halide from Cl to Br or I in RuCl2(CHPh)-
(IMesH2) results in increased initiation rates in ROMP;
nevertheless, propagation rates were found to be reduced
at least in the ROMP of cyclooctene.24 In asymmetric
RCM, an almost dramatic change in enantioselectivity
was found when the ligand sphere in RuCl2(CHPh)-
(NHC)-type complexes bearing chiral NHCs is changed
from Cl to Br and I, respectively.25 Thus, substitution of
the halides in this type of complexes by other groups has
in fact a dramatic effect on reactivity; however, it is hard
to predict whether it will be an increase or decrease since
electronic and steric factors play an important role.
From this set of data, three conclusions can be drawn.
First, the reactivity of soluble, monomeric, both phos-
phine- and NHC-based catalysts bearing electron-with-
drawing ligands in RCM is generally equal or lower than
that of the parent, chlorine-containing systems. Second,
it is evident that with catalysts 1-7 electron-poor double
bonds give lower TONs, whereas electron-rich double
bonds tend to give higher TONs. This finding is in
accordance with the fact that only catalysts containing
electron-withdrawing substituents can in contrast to
their precursor catalysts cyclopolymerize the rather
electron-rich 1,6-heptadiynes.18,21
Finally, reactivity, in particular the differences in
reactivity of the catalysts presented here, clearly under-
lines an early statement given by Fu¨rstner that no
catalyst is an equally good one for every substrate one
can offer, i.e., that reactivity in RCM (and other meta-
thesis based reactions) depends on distinct catalyst-
substrate issues.26
TABLE 2. Results for the RCM of Diethyl
Allyl(2-methylallyl)malonate (12) with Catalysts 1, 2, 6,
and 8-10a
catalyst
mol % of compd
yield (%)
TON
Grubbs 1st
0.2
0.2
0.05
0.2
0.2
0.2
0.2
0.2
0.05
0.2
0.2
0.5
0.1
0.2
83
81
36
99
98
2
410
410
720
500
500
2
Grubbs-Hoveyda 1st
Grubbs 2nd
Grubbs 2nd
Grubbs-Hoveyda 2nd
1
2
3
15
6
84
29
90
0
0
19
32
420
570
450
8
8
9
9
10
10
190
160
a Conditions: solvent, CH2Cl2; T ) 45 °C; t ) 18 h.
Synthesis and Structure of Supported Catalysts
8-10. Numerous immobilized versions of ruthenium-
based metathesis catalysts have been reported so far.27,28
For the purpose of preparing immobilized versions of
catalysts 1-3, hydroxymethylpolystyrene (PS-DVB-CH2-
OH, 1.7 mmol of CH2OH/g, cross-linked with 1% DVB)
was reacted with perfluoroglutaric anhydride following
a procedure published by Nieczypor et al.29,30 and later
optimized by our group (Scheme 3).16-21
TABLE 3. Results for the RCM of
N,N-Diallyl-p-toluenesulfonamide (13) with Catalysts 1, 2,
6, and 8-10
catalyst
mol % of compd
yield (%)
TON
Grubbs 1st
0.02
0.02
0.01
0.02
0.02
0.1
0.02
0.02
0.01
0.05
0.1
52
92
34
54
66
76
59
58
42
91
21
92
< 1
< 1
2600
4600
3400
2700
3300
760
3000
2900
4200
1820
210
180
-
Grubbs-Hoveyda 1st
Grubbs 2nd
Grubbs 2nd
Grubbs-Hoveyda 2nd
1
2
6
(22) Conrad, J. C.; Amoroso, D.; Czechura, P.; Yap, G. P. A.; Fogg,
D. E. Organometallics 2003, 22, 3634-3636.
8
(23) Dias, E. L.; Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc.
1997, 119, 3887-3897.
8
9
(24) Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem. Soc.
2001, 123, 6543-6554.
9
0.5
0.02
0.1
10
10
(25) Seiders, T. J.; Ward, D. W.; Grubbs, R. H. Org. Lett. 2001, 3,
3225-3228.
-
(26) Fu¨rstner, A. Angew. Chem. 2000, 112, 3140-3172.
(27) Buchmeiser, M. R. New. J. Chem. 2004, 28, 549-557.
(28) Yao, Q.; Zhang, Y. J. Am. Chem. Soc. 2004, 126, 74-75.
(29) Nieczypor, P.; van Leeuwen, P. W. N. M.; Mol, J. C.; Lutz, M.;
Spek, A. L. J. Organomet. Chem. 2001, 624, 58-66.
(30) Nieczypor, P.; Buchowicz, W.; Meester, W. J. N.; Rutjes, F. P.
J. T.; Mol, J. C. Tetrahedron Lett. 2001, 42, 7103-7105.
a Conditions: solvent, CH2Cl2; T ) 45 °C; t ) 18 h.
versus DEDAM was low (TON ) 400), and its reactivity
for 14 was actually zero. This is in strong contrast to a
paper by Fogg et al.22 Here, the synthesis of Ru(Py)-
4690 J. Org. Chem., Vol. 70, No. 12, 2005

Ruthenium-Based Metathesis Catalysts
FIGURE 4. 31P-MAS NMR of Ru(PS-DVB-CH2OCOCF2CF2-
CF2COO)(CF3CO2)(PCy3)(dCH-2-(2-PrO)C6H5) (9).
FIGURE 2. 31P-MAS NMR of Ru(PS-DVB-CH2OCOCF2CF2-
CF2COO)(CF3CO2)(PCy3)(IMesH2)(dCHC6H5) (8).
the catalyst loading for 10 was 160 mg/g.20 It is worth
mentioning that catalysts 8 and 9 represent the first
permanently bound versions of both Grubbs’ 2nd and the
Grubbs-Hoveyda 1st generation catalyst. In contrast to
compound 3, the supported version 8 was accessible in
pure analytical form after excessive washing with THF
as evidenced through MAS NMR spectroscopy. Thus, all
supported catalysts were subjected to 1H-, 13C-, and 31P-
MAS NMR spectroscopy. To confirm the structures of the
bound species, both the benzylidene and phosphine
signals were considered. The 31P-MAS NMR spectrum of
8 shows one (broad) phosphine species at δ ) 41.1 ppm
(Figure 2). For comparison, 3 shows two signals at δ )
41.2 and 45.1, the latter resulting from contaminant
AgCl‚PCy3 (δ ) 43.3 ppm in CDCl3). The 13C NMR-CP-
MAS spectrum of 8 shows a signal at δ ) 18.5 ppm
(o-CH3 of the IMesH2-ligand) and shoulder at 21.3 ppm
(p-CH3 of the IMesH2-ligand, Figure 3). The 31P-MAS
NMR of 9 (Figure 4) showed a resonance at δ ) 56.8 ppm,
which is basically the identical shift found in the homo-
geneous mimic (1, δ ) 57.5 ppm). The doublet results
from the coupling of the phosphine with the benzylidene
proton.
FIGURE 3. 13C-CP-MAS NMR of Ru(PS-DVB-CH2-
OCOCF2CF2CF2COO)(CF3CO2)(PCy3)(IMesH2)(dCHC6H5) (8)
(* ) rotating sidebands).
Catalyst loadings of 2.4 and 22.1 mg/g were determined
for catalysts 8 and 9, respectively. As reported previously,
SCHEME 3. Synthesis of Catalysts 8-10
J. Org. Chem, Vol. 70, No. 12, 2005 4691

Halbach et al.
TABLE 4. Results for the RCM of
was consumed, took part in the RCM reactions, and
finally decomposed. Therefore, no recycling studies have
been performed. In the case of supported catalysts,
leaching was low, resulting in an average ruthenium
contamination of the products of 83, 15, and 70 ppb,
respectively, for catalysts 8, 9, and 10.
N-Benzyl-N-but-1-ene-4-ylbut-2-enecarboxylic Amide (14)
with Catalysts 1, 2, and 4-10
catalyst
mol % of compd
yield (%)
TON
Grubbs 1st
0.01
0.1
0.5
0.02
0.1
0.1
0.1
0.5
0.1
0.1
0.1
0.1
0.1
0.02
0.1
0.1
0.5
0.1
4
11
92
58
93
80
<1
8
40
100
Grubbs-Hoveyda 1st
Grubbs-Hoveyda 1st
180
Grubbs 2nd
2900
930
800
Grubbs 2nd
Conclusion
Grubbs-Hoveyda 2nd
1
1
2
4
5
6
7
8
8
9
9
10
In summary, a series of novel supported and unsup-
ported metathesis catalysts have been prepared. No
general trend in terms of reactivity was observed upon
heterogenization. With Hoveyda 1st generation type
catalysts, a significant loss of reactivity was observed.
Both monomeric and polymer-bound catalysts showed
similar behavior. Hoveyda 2nd generation-type catalysts
based on fluorinated carboxylates and phenolates dis-
tinguished between electron-poor substrates and electron-
rich substrates. Lower reactivity was observed with the
former ones, while reactivity appears to be preserved
with the latter. Finally and important enough, im-
mobilization of Grubbs’ 2nd generation catalyst using
fluorinated carboxylates resulted in the preservation,
sometimes even in an increase in RCM reactivity, making
this supported version a highly attractive one for various
application that require catalyst immobilization, e.g., for
high-throughput screening.
16
30
3
3
2
30
20
13
<1
29
99
4
49
<1
130
1450
1000
40
100
a Conditions: solvent, CH2Cl2; T ) 45 °C; t ) 18 h.
TABLE 5. Results for the RCM of
N-Allyl-N-(1-carboxymethyl)but-3-en-1-yl-p-
toluenesulfonamide (15) with Catalysts 1, 2, 6, and 8-10a
catalyst
mol % of compd
yield (%)
TON
Grubbs 1st
0.02
0.1
0.01
0.02
0.1
0.02
0.02
0.02
0.1
0.02
0.1
0.5
0.02
0.1
20
65
31
35
72
28
19
55
98
3
1000
650
1560
1760
720
1400
950
2750
980
150
360
200
70
Grubbs-Hoveyda 1st
Grubbs 2nd
Grubbs-Hoveyda 2nd
Experimental Section
1
2
General Remarks. NMR data were obtained at 300 K at
300.13 (500) MHz for proton, 121.49 (202.48) MHz for phos-
phorus, and at 75.74 (126.23) MHz for carbon, respectively,
in the indicated solvent and are listed in parts per million
downfield from tetramethylsilane for proton and carbon.
Coupling constants are listed in Hz. IR spectra were recorded
using ATR technology. Mass spectra were recorded using EI
(70 eV). Syntheses of the ligands and catalysts were performed
under an argon atmosphere by standard Schlenk techniques
or in an Ar-mediated drybox unless stated otherwise. Reagent-
grade diethyl ether, pentane, THF, and toluene were distilled
from sodium benzophenone ketyl under argon. Reagent-grade
dichloromethane was distilled under argon from calcium
hydride or sicapent (P4O10), respectively. Other solvents and
reagents were used as purchased. Deionized water was used
throughout. RuCl2(dCHPh)(IMesH2)(PCy3) (IMesH2 ) 1,3-bis-
(2,4,6-trimethylphenyl)-4,5-dihydroimidazolin-2-ylidene), RuCl2-
(dCH-2(2-PrO)C6H4)(PCy3), CF3SO3Ag, CF3COOAg, PS-DVB-
CH2OH (100-300 mesh, 1.7 mmol Ar-CH2OH/g, 1% cross-
linked), and perfluoroglutaric anhydride were purchased. A
ruthenium standard containing 1000 ppm of ruthenium (in 1
M HNO3) was used. Catalysts 2, 4, 5, and 10 were prepared
as described in previous disclosures.20,21
MAS Experiments. Samples were packed into 4 mm ZrO2
rotors. Measurements were performed at B0 ) 11.7 T. The
samples were spun at 15 kHz unless stated otherwise. 1H NMR
spectra were recorded as the sum of 100 scans with a recycle
time of 2 s. A 2.8 µs pulse was applied. The RF field was 89
kHz. Spectra were calibrated against an external standard of
adamantane (2.00 ppm). 13C-CP-MAS NMR was measured
using cross-polarization with gradient pulses. The contact time
was 2 ms. Continuous wave decoupling was applied during
data acquisition. The spectra consisted of at least 10 000 scans
with a recycle rate of 5 s. For calibration, the peak of the CH
groups in adamantane was set to 29.472 ppm. To recognize
spinning sidebands, a second spectrum was recorded at a
spinning rate of 12 kHz, so that spinning sidebands appeared
with different shifts. For 31P NMR spectra, 12000 scans were
6
8
8
9
9
36
100
1
9
10
10
8
80
a Conditions: solvent, CH2Cl2; T ) 45 °C; t ) 18 h.
Reactivity of Supported Catalysts 8-10 in RCM.
All three supported catalysts were investigated for their
RCM reactivity. While the use of the supported catalysts
9 and 10 in the RCM of diethyl diallylmalonate (DEDAM,
12), diethyl allyl(2-methylallyl)malonate (11), N,N-dial-
lyl-p-toluenesulfonamide (14), N-benzyl-N-but-1-en-4-
ylbut-2-enecarboxylic amide (13), and N-allyl-N-(1-
carboxymethyl)but-3-en-1-yl-p-toluenesulfonamide (15)
resulted in TONs up to 360 (Tables 1-5), the reactivity
of 3 was dramatically increased by immobilization (re-
sulting in 8). In fact, reactivity in some cases rivaled and
exceeded reactivity of the precursor catalysts (Tables
3-5). We attribute this to two different effects. On one
hand, catalyst reformation via phosphine exchange is
certainly faster than it is the case with 9 and 10, which
require metathesis with the intermediary formed 2-(2-
PrO)styrene. On the other hand, bimolecular decomposi-
tion is certainly more effectively suppressed via immo-
bilization, leading to improved stability and durability
compared to parent 3.
Recyclability and Leaching. Reactions described in
this paper were run to an extent where no unreacted
catalyst could be recovered. This implies that all catalyst
4692 J. Org. Chem., Vol. 70, No. 12, 2005

Ruthenium-Based Metathesis Catalysts
recorded with a recycle time of 5 s. A 4 µs pulse was used.
The RF field was 63 kHz. Chemical shifts for phosphorus are
reported relative to (NH4)H2PO4 at 1.11 ppm.
CHOAr, J ) 6.0), 3.98 (s, 4 H, N(CH2)2N), 2.36 (s, 6 H, mesityl
p-CH3), 2.03 (s, 12 H, mesityl o-CH3), 0.98 (d, 6 H, (CH3)2-
CHOAr, J ) 6.0). 13C{1H}-NMR (CDCl3): δ 313.5, 211.6, 160.7,
153.7, 144.8, 143.8, 139.2, 134.8, 129.6, 129.5, 128.6, 126.5,
123.5, 122.6, 111.0, 74.4, 51.3, 21.1, 20.0, 17.9. ESI-MS
(acetonitrile/water) calcd for C45H38N2F10O5Ru: 978.17, found
782.9 (M+• + H+ - C6F5CO, 33), 307.4 (IMesH2 + H+, 29). Anal.
Calcd for C45H38N2F10O5Ru: C, 55.21; H, 3.92; N, 2.86.
Found: C, 54.93; H, 4.30; N, 2.53.
RCM Experiments. Reactions were run for 18 h in order
to achieve complete conversion. For the homogeneous catalysts,
a carousel working station under N2 was used. The reaction
volume was 1-4 mL. Heterogeneous catalysts were run in a
N2-mediated drybox using microreactors in a thermo shaker.
Analyses were done by NMR after filtration for catalysts 2, 4,
and 6 and by HPLC for catalysts 8 and 10. No other products
were found.
Ru(C6F5O)2(IMesH2)(dCH-2-(2-PrO)C6H4) (7). RuCl2-
(IMesH2)(dCH-2-(2-PrO)C6H4) (200 mg, 0.32 mmol) was dis-
solved in 10 mL of THF, and a solution of TlOC6F5 (2 equiv,
260 mg, 0.64 mol), generated in situ from TlOC2H5 and C6F5-
OH in a mixture of 1 mL of THF, 1 mL of methanol, and 1 mL
of toluene, was slowly added. Stirring was continued for 3 h.
A color change from red to brown and the formation of a
precipitate was observed. The precipitate was filtered off and
the solution evaporated to dryness. The solid was redissolved
in CH2Cl2, centrifuged, and filtered through Celite. Drying in
vacuo provided a brown powder (249 mg, 0.27 mmol, 85%).
IR-ATR (cm-1): 2966 (br, s), 2921 (br, s), 1644 (w), 1617 (w),
1591 (w), 1575 (w), 1495 (vs), 1455 (vs), 1379 (w), 1294 (w),
1260 (vs), 1203 (w), 1158 (s), 1112 (s), 1106 (vs), 980 (vs), 851
(s), 794 (vs), 741 (vs), 629 (w). 1H NMR (CDCl3): δ 17.10 (s, 1
H, RudCHAr), 7.44 (dd, 1 H, aromatic CH, J1 ) 1.5, J2 ) 2.0),
7.11 (s, 4 H, mesityl CH), 7.26 (dd, 1 H, aromatic CH, J1 )
7.3, J2 ) 7.8), 6.97 (dd, 1 H, aromatic CH, J1 ) 1.5, J2 ) 1.5),
6.80 (d, 1 H, aromatic CH, J ) 8.3), 3.82 (sep, 1 H, (CH3)2-
CHOAr, J ) 6.0), 4.10 (s, 4 H, N(CH2)2N), 2.41 (s, 6 H, mesityl
p-CH3), 2.28 (s, 12 H, mesityl o-CH3), 0.67 (d, 6 H, (CH3)2-
CHOAr, J ) 6.0). 13C{1H}-NMR (CDCl3): δ 296.3, 210.4, 150.2,
143.3, 142.3, 139.3, 136.1, 129.5, 128.7, 127.8, 126.5, 123.3,
122.0, 111.4, 74.7, 51.2, 21.2, 19.3, 18.1. HRMS calcd for
C43H38N2F10O3Ru: 922.18, found 922.30.
Ru(CF3COO)2(PCy3)(dCH-2-(2-PrO)C6H4) (1). Silver tri-
fluroacetate (23.2 mg, 0.105 mmol) was suspended in 10 mL
of CH2Cl2. RuCl2(PCy3)(dCH-2-(2-PrO)C6H4) (30 mg, 0.05
mmol), dissolved in 0.5 mL of CH2Cl2, was added, and the
solution was stirred for 30 min in the absence of light. The
mixture was filtered through Celite/cellulose, and the solvent
was reduced in vacuo. The product (37 mg, 0.049 mmol, 98%)
was obtained as a red, foamy solid. Recrystallization from
hexane/CH2Cl2 yielded red needles. IR (cm-1): 724 (m), 735
(m), 863 (m), 929 (w), 1144 (s), 1190 (ss), 1391 (w), 1453 (m),
1642 (m), 1712 (s), 2854 (m), 2931 (s). 1H NMR (500 MHz;
CDCl3): δ 1.22-1.33 (m, 9H), 1.45 (d, J ) 6.7, 6H), 1.48-1.58
(m, 6H), 1.67-1.76 (m, 3H), 1.78-1.95 (m, 12H), 2.02-2.12
(m, 3H), 4.96-5.05 (m, 1H), 7.03 (d, J ) 8, 1H), 7.19 (dd, J1 )
J2 ) 7.5, 1H), 7.62 (dd, J1 ) J2 ) 8, 1H), 7.80 (d, J ) 7.5, 1H),
18.39 (d, J ) 6, 1H). 13C NMR (125.8 MHz; CDCl3): δ 21.1,
26.3, 27.7, 29.0, 34.2, 34.4, 76.7, 112.7, 113.8 (q, J ) 288.5),
123.4, 124.4, 131.2, 144.2 (q), 155.22 (q), 163.48 (q, J ) 33.7),
308.5 (d, J ) 15.1). 31P NMR (202.5 MHz; CDCl3): δ 56.6. 19
F
NMR (470.6 MHz; CDCl3): δ -74.6. EI-MS (200 °C): m/z )
756 (M+), 642 (46), 600 (6), 404 (10), 375 (10), 281 (24), 243
(100%), 198 (54), 120 (62), 91 (36), 83 (54), 78 (60), 55 (75).
HR-MS calcd for C32H45F6O5PRu): 756.1946, found 756.1933.
Ru(PS-DVB-CH2OCOCF2CF2CF2COO)(CF3CO2)-
(dCHC6H5)(IMesH2)(PCy3) (8). PS-DVB-CH2OCOCF2CF2-
CF2COOAg was prepared according to the literature.31 PS-
DVB-CH2OCOCF2CF2CF2COOAg (500 mg) was suspended in
CH2Cl2 (15 mL), and RuCl2(dCHC6H5)(IMesH2)(PCy3) (80 mg,
0.094 mmol), dissolved in CH2Cl2 (2 mL), was added slowly.
The mixture was stirred for 90 min in the absence of light,
and CF3COOAg (21 mg, 0.094 mmol), dissolved in 2 mL of
THF, was added. The reaction mixture was stirred for a
further 90 min. The product was filtered and washed three
times with THF (10 mL) until the filtrate was colorless. Drying
in vacuo provided a red-lilac powder (450 mg). IR-ATR (cm-1):
3082 (w), 3058 (w), 3024 (s), 2918 (br), 2849 (w), 1869 (w),
1773 (br), 1660 (br), 1601 (s), 1510 (s), 1492 (s), 1308 (w), 1265
Ru(CF3COO)2(dCHC6H5)(IMesH2)(PCy3) (3). CF3COOAg
(10 mg, 0,043 mmol) was suspended in 10 mL of CH2Cl2 and
stirred for 5 min. A solution of RuCl2(dCH-C6H5)(IMesH2)-
(PCy3) (17.8 mg, 0.021 mmol) in 2 mL of CH2Cl2 was added.
Stirring was continued for 5 min. A color change from red to
green-brown and the formation of a precipitate was observed.
The precipitate was filtered off, and the solution was filtered
through Celite. Drying in vacuo provided a red powder. No
satisfying elemental analysis was obtained due to coprecipi-
tation of AgCl‚PCy3. Yield: 18.5 mg (0.018 mmol, 85%). IR-
ATR (cm-1): 2925 (s), 2851 (s), 1686 (s), 1631 (s), 1482 (s),
1448 (s), 1265 (s), 1192 (vs), 1136 (vs), 1030 (w), 1003 (w), 962
(vw), 851 (s), 785 (w), 737 (vs), 695 (s), 618 (vw). 1H NMR
(CDCl3): δ 18.34 (s, 1 H, RudCH), 7.45 (d, 1 H, J ) 5.5), 7.27
(m, 4 H), 6.93 (d, 1H, J ) 6.3), 6.77 (m, 2 H), 4.09 (s, 4 H),
2.17 (m, 17 H), 1.80 (s, 20 H), 1.19 (m, 20 H). 13C NMR
(CDCl3): δ 289.2, 184.4, 159.6, 140.7, 137.3, 135.4, 134.7, 129.7,
128.7, 127.8, 52.0, 31.9, 29.0, 27.8, 26.2, 20.8, 17.7, 14.1.
Additional signals for AgCl‚PCy3: 31.2, 27.1, 25.8. 31P{1H}-
NMR (CDCl3): δ ) 41.2, 45.1 (AgCl‚PCy3).
1
(s), 1154 (br), 1029 (br), 908 (w), 819 (w), 751 (s), 697 (s). H-
MAS NMR: δ 6.99 (br), 3.79 (sh), 1.95 (sh). 13C-CP-MAS
NMR: δ 18.5, 21.3 (sh), 26.0, 40.6, 45.7 (sh), 65.1, 68.1, 128.1,
139.4, 145.6. 31P-MAS NMR: δ 41.1. ICP-OES: 0.22 mg Ru/
g. Anal. Found: C, 75.24; H, 6.45; N, 0.47.
Ru(PS-DVB-CH2OCOCF2CF2CF2COO)(CF3CO2)(dCH-
2-(2-PrO)C6H4)(PCy3) (9). PS-DVB-CH2OCOCF2CF2CF2-
COOAg (850 mg) was suspended in CH2Cl2 (25 mL), and
RuCl2(dCH-2-(2-PrO)C6H4)(PCy3) (95.5 mg, 0.152 mmol) in
CH2Cl2 (2 mL) was added. The mixture was stirred for 90 min
in the absence of light. CF3COOAg (35.2 mg, 0.152 mmol) was
dissolved in THF (2 mL) and added to the catalyst. Stirring
was continued for another 90 min. The product was filtered
and washed three times with THF (10 mL) until the filtrate
was colorless. Drying in vacuo provided a red-lilac powder (650
mg). IR (cm-1): 3082 (w), 3058 (w), 3024 (s), 2919 (br), 2849
(w), 1939 (w), 1773 (br), 1656 (br), 1510 (s), 1451 (s), 1308 (w),
1154 (br), 1029 (br), 907 (w), 819 (w), 749 (s), 697 (s). 1H-MAS
NMR δ 7.34 (br), 4.05 (sh), 2.00 (sh). 13C-CP-MAS NMR δ 40.7,
45.2 (sh), 65.1, 68.1, 128.0, 146.0. 31P-MAS NMR δ 56.81 (d).
ICP-OES: 2.62 mg Ru/g. Anal. Found: C, 73.55; H, 6.36.
Ru(C6F5CO2)2(IMesH2)(dCH-2-(2-PrO)C6H4) (6). RuCl2-
(IMesH2)(dCH-2-(2-PrO)C6H4) (200 mg, 0.32 mmol) was dis-
solved in 10 mL of THF, and a solution of TlCO2C6F5 (2 equiv,
266 mg, 0.64 mmol), generated in situ from TlOC2H5 and C6F5-
CO2H in a mixture of 1 mL of THF, 1 mL of methanol, and 1
mL of toluene, was slowly added. Stirring was continued for 3
h. A color change from red to brown and the formation of a
precipitate was observed. The precipitate was filtered off and
the solution evaporated to dryness. The solid was redissolved
in CH2Cl2, centrifuged, and flashed over Celite. Drying in
vacuo provided a lilac powder (272 mg, 0.28 mmol, 87%). IR
(cm-1): 2924 (br), 1663 (vs), 1594 (w), 1517 (s), 1479 (vs), 1333
(vs), 1267 (vs), 1210 (w), 1098 (s), 1034 (w), 987 (vs), 920 (s),
1
878 (w), 842 (w), 745 (vs), 699 (s), 574 (s). H NMR (CDCl3):
δ 17.33 (s, 1 H, RudCHAr), 7.43 (d, 1 H, aromatic CH, J )
9.0), 7.01 (s, 4 H, mesityl CH), 7.23 (dd, 1 H, aromatic CH, J1
) 7.5, J2 ) 9.0), 6.83 (dd, 1 H, aromatic CH, J1 ) 7.5, J2 7.3),
6.59 (d, 1 H, aromatic CH, J ) 8.3), 4.57 (sep, 1 H, (CH3)2-
(31) Yang, L.; Mayr, M.; Wurst, K.; Buchmeiser, M. R. Chem. Eur.
J. 2004, 10, 5761-5770.
J. Org. Chem, Vol. 70, No. 12, 2005 4693

Halbach et al.
Determination of Ruthenium Content. Aqua regia (3.0
mL) was added to the combined effluents from which the
solvent was removed. The mixture was placed inside high-
pressure Teflon tubes, and leaching was carried out under
microwave conditions (50, 600, and 450 W pulses, respectively,
t ) 32 min). After cooling to room temperature, the mixture
was filtered, and water was added up to a volume of 10.00
mL.
carried out with Ru standards containing 0, 5, and 10 ppm of
Ru.
Acknowledgment. Financial support provided by
the Austrian Science Fund (FWF Vienna, Project No.
Y-158) and the Freistaat Bayern is gratefully acknowl-
edged. D.F. and S.M. thank the Graduiertenkolleg
“Synthesische, mechanistische und reaktionstechnische
Aspekte von Metallkatalysatoren” for a stipend.
Ru Measurements. Ru was measured by ICP-OES (λ )
240.272 nm, ion line). The background was measured at λ )
240.287 and 240.257 nm, respectively. Standardization was
JO0477594
4694 J. Org. Chem., Vol. 70, No. 12, 2005