Preparation of tungsten-based olefin metathesis catalysts supported on alumina

Article abstract of DOI:10.1002/adsc.201100200

A new tungsten alkylidene complex, W(NAr)(CHCMe₂Ph)(OHIPT- NMe₂)(pyrrolide) {Ar=2,6-(i-Pr)₂C₆H ₃; HIPT-NMe₂=2,6-[2,4,6-(i-Pr)₃C ₆H₂]₂-4-NMe₂-C₆H ₂}, has been synthesized and shown to be highly selective for Z homocoupling metathesis of selected terminal olefins in pentane, as is W(NAr)(CH₂CH₂CH₂)(OHIPT)(pyrrolide) (5). Both 5 and W(NAr)(CHCMe₂Ph)(OHIPT-NMe₂)(pyrrolide) (6) are adsorbed onto calcined alumina. Control experiments and metathesis homocoupling of four substrates lead to the conclusions that 5 is largely adsorbed in a reaction that liberates HIPTOH, while 6 is adsorbed largely through an interaction between the dimethylamino group and an acidic site on the surface. There is no evidence that any adsorbed catalyst can give rise to Z selectivity of a magnitude equal to that found in a homogeneous reaction involving 5 or 6. Copyright

Full text of DOI:10.1002/adsc.201100200

FULL PAPERS
DOI: 10.1002/adsc.201100200
Preparation of Tungsten-Based Olefin Metathesis Catalysts
Supported on Alumina
Jian Yuan,a Erik M. Townsend,a Richard R. Schrock,a,* Alan S. Goldman,b,*
Peter Mꢀller,a and Michael K. Takasea
a
Department of Chemistry 6-331, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA
Fax : (+1)-617-253-7670; e-mail: rrs@mit.edu
Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, Piscataway, New Jersey
b
08854, USA
Fax : (+1)-732-445-5312; e-mail: alan.goldman@rutgers.edu
Received: March 21, 2011; Revised: May 16, 2011; Published online: July 14, 2011
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adcs.201100200.
Abstract:
(NAr)(CHCMe2Ph)(OHIPT-NMe2)(pyrrolide)
{Ar=2,6-(i-Pr)2C6H3;
A new tungsten alkylidene complex, is largely adsorbed in a reaction that liberates
W
N
G
HIPTOH, while 6 is adsorbed largely through an in-
HIPT-NMe2 =2,6-[2,4,6-(i- teraction between the dimethylamino group and an
Pr)3C6H2]2-4-NMe2-C6H2}, has been synthesized and acidic site on the surface. There is no evidence that
shown to be highly selective for Z homocoupling any adsorbed catalyst can give rise to Z selectivity of
metathesis of selected terminal olefins in pentane, as a magnitude equal to that found in a homogeneous
is
Both
W
N
ACHTUNGTRENNUNG
5
W
N
ACHTUNGTRENNUNG
NMe2)(pyrrolide) (6) are adsorbed onto calcined alu-
mina. Control experiments and metathesis homocou- Keywords: alkylidene species; alumina; olefin meta-
pling of four substrates lead to the conclusions that 5 thesis; supported catalysts; tungsten
Introduction
supported catalysts are not known. Any chemical at-
tachment of the metal to a surface through an oxygen
In the last dozen years various “well-defined” automatically alters the essential nature of the cata-
Mo/W[1] or Ru[2] olefin metathesis catalysts have been lyst from what it is in solution, where ligands have
attached to or prepared on a solid support, either an been finely tuned for a solution environment. Any se-
organic polymer such as polystyrene[3] or a “hard” lectivity that is present for the homogeneous catalyst
support such as silica, alumina, or molecular sieves.[4] therefore is unlikely to be found for the supported
The ultimate goal is to produce a heterogeneous ver- catalyst if that selectivity depends upon a specific cat-
sion of a successful homogeneous catalyst, and per- alyst structure.
haps one that is longer-lived and recyclable. A “hard”
Approaches to supporting iridium catalysts that are
support is required in order to avoid problems associ- employed in Ir/Mo-catalyzed “alkane metathesis” re-
ated with swelling of a support such as polystyrene in actions have been explored recently.[5] One approach
a solvent, which is intolerable if construction of chro- involved attaching a basic functional group (e.g., di-
matography columns is a goal.
methylamino) in the para-position of the phenyl ring
Syntheses of Mo/W catalysts on hard supports are in a “PCP pincer” ligand in order to link the Ir com-
rare.[4] Syntheses usually consist of a reaction between plex to acidic surface sites on calcined g-alumina
M(NR)ACHTUNGTRENNUNG(CHR’)ACHTUNGTRENNUNG(OR’’)2 or M(NR)ACHTNUGTERN(NUGN CHR’)(pyrrolide)2 through a Lewis acid/base interaction. In the process
(M=Mo or W) species and some OH group on the of evaluating Ir catalysts supported on alumina it was
support to yield R’’OH or pyrrole, respectively. With noted that functional alkane metathesis catalysts
the exception of catalysts prepared in a reaction be- could be prepared simply through addition of
tween an M(NR)
ACHTUNGTRENNUNG(CHR’)(pyrrolide)2 complex and de- MoACUTHNGTRENNU(GN NAr)ACHTUNRTGE(NNGNU CHCMe2Ph)ACTHNUGTNERN[UNG OCMeACTHUNTGRENU(GN CF3)2]2 to the alumi-
hydroxylated silica,[4d] the structures of the resulting na-supported Ir complex and that both the Ir and the
Adv. Synth. Catal. 2011, 353, 1985 – 1992
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1985

Jian Yuan et al.
FULL PAPERS
Scheme 1.
Mo catalysts appeared to be adsorbed completely g-alumina other than the Lewis acid/base interaction
onto g-alumina to give active species that function in between the dimethylamino group and an acidic sur-
tandem to metathesize alkanes. It also has been noted face site could give rise to a metathesis catalyst of un-
that Mo-based bisalkoxide metathesis catalysts can be known type that would most likely not produce a cat-
adsorbed on silica-alumina[4b] or molecular sieves.[4c] alyst with the same Z-selectivity as the one in solu-
Supporting a catalyst through a Lewis acid/base inter- tion. Sensitive Z-selective reactions such as ROMP
action in a position relatively remote from the metal polymerization[7a] or coupling of terminal olefins[7b]
is an attractive way to increase the chances that the could serve as an indicator of whether the selectivity
supported catalyst will behave like its homogeneous has been maintained on the solid support. We report
analog.
here an evaluation of approaches to metathesis cata-
We have developed Mo and W monoaryloxide pyr- lysts supported on alumina and an evaluation of their
rolide (or MAP) catalysts that contain the sterically efficiency for homocoupling a small selection of sub-
demanding 2,6-[2,4,6-(i-Pr)3C6H2]2C6H3O (HIPTO)[6] strates.
aryloxide ligand,[7] among others.[8] The HIPTO ligand
allows W catalysts to homocouple terminal olefins to
give Z-internal olefins. We decided to attempt to pre- Results and Discussion
pare MAP species in which a dimethylamino group is
attached in the para position of the HIPTO ligand, Synthesis of Ligands and Catalysts
and to attach an intact, Z-selective homogeneous cat-
alyst to g-alumina through the dimethylamino group The desired phenol HOHIPT-NMe2 (4) was prepared
to yield a heterogeneous version of the homogeneous as shown in Scheme 1. Treatment of hexaisopropylter-
catalyst. Any reaction between such a MAP species phenol with bromine in acetic acid gave 1 in 71%
[M(NR)ACHTUNGTRENNUNG(CHCMe2Ph)(OHIPT-NMe2)(pyrrolide)] and yield. Protection of the phenol with a MOM group to
1986
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1985 – 1992

Preparation of Tungsten-Based Olefin Metathesis Catalysts Supported on Alumina
give 2 (98% yield) was followed by palladium-cata-
lyzed cross-coupling of 2 with dimethylamine to give
3 in 72% yield. Deprotection of 3 with HCl gave
HOHIPT-NMe2 (4) in 82% yield. The overall yield of
4 is ~40%.
The reaction between
ACHTUNGTRENNUNG(DME)ACHTUNGTRENNUNG(Pyr)2
W
G
E
[9]
W
N
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
A
R
ACHTUNGTRENNUNG
a pentane solution of 6 was treated with one atmos-
phere of ethylene. The expected metallacyclobutane
complex, WACHTUNGTRENNUNG(NAr)(CH2CH2CH2)(OHIPT-NMe2)CAHTUNGTRNE(NUGN Pyr)
(7a) and CH2=CHCMe2Ph were observed in solution
(in NMR studies), but removal of ethylene from the
crude mixture led to reformation of 6 along with the
b-substituted tungstacyclobutane complex, W
[CH2CH(CMe2Ph)CH2](OHIPT-NMe2)(Pyr) (7b); 7b
formed as a consequence of a back reaction between
(NAr)(CH2)(OHIPT-NMe2)(Pyr) and
CH2=CHCMe2Ph as ethylene was removed. Com-
pound 5’ was prepared from 5 and neat CH2=CH-t-Bu
(see Experimental Section) in order to carry out con-
trol experiments, as described later.
The crystal chosen for an X-ray structural study of
7a turned out to be a cocrystallized mixture of 6
(24%) and 7b (76%); there were no significant differ-
ences (disorder) in the aryl oxide, pyrrolide, and
imido ligands between the two structures. Both struc-
tures could be solved and are shown in Figure 1 and
Figure 2. A complete list of bond lengths and angles
can be found in the Supporting Information.
ACHTUNGTRNE(NUNG NAr)-
Figure 1. Thermal ellipsoid drawing of the non-hydrogen
atoms of 6, showing the partial atom labeling scheme. Ther-
mal ellipsoids are shown at the 30% level. Selected bond
G
E
ACHTUNGTRENNUNG
W
R
G
ACHTUNGTRENNUNG
À
À
distances (ꢂ) and angles (8): W1 C1A=1.947(17); W1
À
À
À
N2=1.735(5); W1 N1=1.962(17); W1 O1=1.880(3); W1
C1A C2A=144.6(13); N2 W1 C1A=103.1(5); O1 W1
À
À
À
À
À
À
À
À
À
C1A=108.6(5); N2 W1 N1A=105.2(10); O1 W1 N1A=
À
À
À
À
110.5(9);
123.47(14);
165.6(3).
C1A W1 N1A=104.1(8);
N2 W1 O1=
W1 O1 C41=
À
À
À
À
C21 N2 W1=179.3(3);
The structure of 6 is typical of MAP species in that
the alkylidene is syn and the pyrrolide ligand is
bound in a h1 fashion. (In a syn isomer the alkylidene
substituent points toward the imido ligand.) The
greater stability of syn isomers versus anti isomers[8a]
is ascribed to stabilization provided through an agos-
tic interaction of the CHa electrons with the metal
center.[1] The dimethylamino group in 6 is in the para
position of the central phenyl ring, as expected.
Unlike known unsubstituted molybdacyclobutane[8d]
and tungstacyclobutane[7a] MAP species, which are es-
sentially trigonal bipyramids, 7b is closest to a square
pyramid (SP) with the imido as the apical ligand (t
value[10] =0.060). Compound 7b is the first substituted
metallacyclobutane MAP species to be crystallo-
graphically characterized. The metallacyclobutane
Figure 2. Thermal ellipsoid drawing of the non-hydrogen
atoms of 7b, showing the partial atom labeling scheme.
Thermal ellipsoids are shown at the 30% level. Selected
À
À
ring is bent, with the angle between the C1 W1 C3
À
À
bond distances (ꢂ) and angles (8): W1 C1=2.121(5); W1
À
À
and C1 C2 C3 planes being 36.1(4)8, as found in b-
substituted SP bisalkoxide metallacyclobutane com-
À
À
À
C3=2.190(5); W1 C2=2.748(5); C1 C2=1.537(8); C2
À
À
À
C3=1.517(8); W1 N2=1.735(3); W1 N1=2.053(6); W1
plexes.[11] The interior angles of the MC3 ring [C1
W1 C3=62.8(2); C1 C2 C3=94.7(4); W1 C3 C2=
93.7(3); W1 C1 C2=95.9(3)] are similar to the inte-
rior angles in SP bisalkoxide metallacyclobutane com-
À
À
À
À
À
O1=1.880(3); C1 W1 C3=62.8(2); C1 C2 C3=94.7(4);
À
À
À
À
À
À
À
À
À
À
À
W1 C3 C2=93.7(3); W1 C1 C2=95.9(3); C1 W1 N1=
À
À
À
À
À
À
137.8(3);
C3 W1 O1=134.19(17);
N2 W1 O1=
À
À
À
À
123.47(14);
C21 N2 W1=179.3(3);
W1 O1 C41=
plexes, and differ from those observed in TBP metal- 165.6(3).
Adv. Synth. Catal. 2011, 353, 1985 – 1992
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1987

Jian Yuan et al.
FULL PAPERS
Table 1. Screening of substrates.[a]
5
5/Al2O3
6
6/Al2O3
Z
Conv.
Z
Conv.
Z
Conv.
Z
Conv.
1-octene
94
96
93
93
69
33
3
0 (47[b])
78
83
0 (25[b])
14
5
93
93
95
91
57
42
4
30
47
23[c]
3
CH2=CHCH2B(pinacolate)
CH2=CHCH2OCH2Ph
CH2=CHCH2Ph
91[c]
94
93
31
94
66
31
62
[a]
All data are listed in %. See Supporting Information for all experimental details.
In C6D6 at 808C after 3 h.
Average of four independent runs (% Z/% conv.=88/23; 90/25; 91/21; 93/22).
[b]
[c]
lacycles (either bisalkoxides or MAP species). It (Table 1) as under different conditions reported else-
should be noted that the success of Z-selective meta- where.[7b] We ascribe the low conversions (3–4%) for
thesis by WACHTUNGTRENNUNG(OHIPT) complexes is proposed to be a allyl benzyl ether to inhibition through binding of the
consequence of the inability of any substituent on a ether oxygen to the metal, most likely intramolecular-
TBP metallacycle intermediate to point away from ly in a W(CHCH2OCH2Ph) intermediate. The slightly
the imido ligand. However, in 7b the b substituent is reduced conversions in the case of CH2=CHCH2B-
pointed away from the imido ligand, in part since the
ACHTUNGTREN(NUNG pin) could be ascribed to a related binding of an
metallacyclobutane ring is bent, but also because the oxygen in the substrate to the metal.
structure is not a trigonal bipyramid. It is unclear
Before attempting to explain the results shown in
what this means in terms of stereoselectivity in meta- Table 1, experiments were performed in order to de-
thesis, since TBP species are proposed to be closer to termine how much, if any, intact catalyst is removed
the alkylidene/olefin intermediate in a metathesis pro- from the alumina surface when 5/Al2O3, 5’/Al2O3, and
cess than a SP metallacycle, and TBP and SP species 6/Al2O3 are washed with pentane, ether, or dichloro-
readily interconvert.[11] These and other issues sur- methane (see Supporting Information for details).
rounding the details of the metathesis process will be The amounts of 5 or 6 in the washes were quantified
discussed in due course elsewhere.
versus an internal standard in proton NMR studies as
shown in Table 2. Greater than 90% 6 could be re-
moved from 6/Al2O3 with ether and dichloromethane,
but virtually none was removed in pentane. These re-
sults suggest that the dimethylamino group in 6 serves
Synthesis and Reactions of Supported Catalysts
A solution of 6 in pentane was added to a stirred sus- as the primary site for binding to alumina in 6/Al2O3,
pension of calcined (at 5008C) neutral g-alumina in that this binding mode is much preferred over any
pentane. Over a period of 10 min, the yellow solution other mode of binding or reaction of 6 with the sur-
became colorless as 6 was adsorbed on alumina to face, that the catalyst is not altered when it binds to
give 6/Al2O3 (0.05 mmol tungsten/g alumina, 95% the surface, and that the catalyst is removed from the
yield). The same procedure with 5 yielded 5/Al2O3. surface by ether or dichloromethane. In contrast, no
Compound 5’ was also adsorbed onto g-alumina to detectable 5 could be washed from 5/Al2O3 with any
yield 5’/Al2O3.
of the three solvents and HIPTOH was liberated
The Z-selective homometatheses of four substrates from 5/Al2O3 (approximately one equivalent in ether
(Table 1) were employed as a means of interrogating and dichloromethane). We propose that addition of 5
the nature of any catalyst that is supported on alumi- to alumina largely leads to reaction of 5 with one or
na or present in solution. All reactions were carried more surface sites to give one or more metal-contain-
out in 1 mL pentane at room temperature for 16 h ing surface species which may or may not be metathe-
using 0.001 mmol W catalyst and 0.025 mmol of sub-
strate (4 mol% W catalyst). The most telling datum
will be the % Z-product; in a reaction involving a
supported catalyst, conversion can depend on small
Table 2. Catalyst recovery upon washing supported catalysts
with solvent.
variations in experimental details, primarily the effi-
ciency of dispersing the supported catalyst in solution
through stirring.
5/Al2O3
5’/Al2O3
6/Al2O3
Pentane
Ether
Dichloromethane
~0%[a]
~0%[b]
~0%[b]
~0%[a]
~0%
91%
93%
~0%[b]
In homogeneous metathesis homocoupling reac-
tions, 6 showed high Z-selectivity for all four sub-
strates (~94% Z-olefin product; Table 1). The same
(not determined)
[a]
Approximately 0.6 equiv. of free HIPTOH recovered.
Approximately 1.0 equiv. of free HIPTOH recovered.
[b]
results were obtained employing
5 in pentane
1988
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1985 – 1992

Preparation of Tungsten-Based Olefin Metathesis Catalysts Supported on Alumina
Table 3. 1-Octene homocoupling in pentane/ether mixtures.
“minor” steric interactions between intermediates and
the surface.
Entry Catalyst Solvent[a]
Product
The effect of diethyl ether on the metathesis of 1-
octene is shown in Table 3. (entries 1, 6, 11, and 16
are identical to entries in Table 1.) The % Z decreases
as ether is added to homogeneous reactions employ-
ing 5 or 6. We propose that a decrease in % Z prod-
uct is a consequence of some relatively non-selective
pathway becoming competitive in the presence of
ether; its exact nature is not known. Reactions that
employ 5/Al2O3 begin to show conversion as ether is
added (entries 6–9). Either some unknown type of
catalyst is being removed from the surface, or ether
somehow activates the surface bound species in 5/
Al2O3; neither is likely to give rise to a high Z-prod-
uct. In run 10 the % Z is 72%, but conversion is low;
therefore traces of highly active 5 are likely to be
washed from the surface under these conditions. Ad-
dition of ether to reactions that employ 6/Al2O3 in-
creases conversion and the % Z rises to be approxi-
mately the same in ether as for 6 itself (92–93% and
80–85%, respectively; runs 20 and 15). These data
suggest that ether is removing some 6 from the sur-
face, but that some other species, possibly surface-
bound 6, remains active, and has a much poorer Z-se-
lectivity (and probably also activity) than homogene-
ous 6.
% Z % Conv.
1
2
3
4
5
6
7
8
5
5
5
5
5
pentane
94
69
98
48
55
59
0
pentane (1 equiv. ether) 95
pentane (10 equiv. ether) 95
pentane (50 equiv. ether) 80
ether
67
na
5/Al2O3 pentane
5/Al2O3 pentane (1 equiv. ether) 47
5/Al2O3 pentane (10 equiv. ether) 49
5/Al2O3 pentane (50 equiv. ther) 57
12
42
20
3
9
10
11
12
13
14
15
5/Al2O3 ether
72
6
6
6
6
6
pentane
93
57
92
87
68
64
46
47
70
87
71
61
81
15
13
pentane (1 equiv. ether) 90
pentane (10 equiv. ether) 93
pentane (50 equiv. ether) 87
ether
ether
85
80
30
16
17
18
19
20
6/Al2O3 pentane
6/Al2O3 pentane (1 equiv. ether) 38
6/Al2O3 pentane (10 equiv. ether) 50
6/Al2O3 pentane (50 equiv. ether) 67
6/Al2O3 ether
ether
5’/Al2O3 pentane
92
93
47
21
22
5’/Al2O3 pentane (50 equiv. ether) 64
[a]
Equivalents of ether are relative to 1-octene.
Results similar to those shown in Table 3 were ob-
served for the metathesis coupling of allylbenzene.
sis active. In the reaction between 5 and alumina, the This table is provided in the Supporting Information.
liberated HIPTOH remains bound to the surface until Catalyst 6/Al2O3 yielded a relatively high (91%) %
completely removed by washing, most efficiently with Z-product from CH2=CHCH2B(pin) (Table 1), al-
AHCTUNGTRENNUNG
ether and dichloromethane. Similar results were ob- though conversion is only 23%. After stirring a sus-
served for 5’/Al2O3 as for 5/Al2O3, thus providing evi- pension of 6/Al2O3 in pentane for 16 h, the mixture
dence that the 5 and 5’ react similarly with Al2O3, and was filtered to give a colorless filtrate and yellow
therefore that comparison of 5/Al2O3 with 6/Al2O3 is solid. The colorless filtrate showed no activity for
valid.
The lack of any conversion of 1-octene to product the yellow solid (recovered 6/Al2O3) gave 20% prod-
with 5/Al2O3 (Table 1) in pentane at 228C suggests uct (91% Z) from CH2=CHCH2B(pin) in fresh pen-
metathesis homocoupling of CH2=CHCH2BACTHNGUTRENN(UG pin), but
AHCTUNGTRENNUNG
that at room temperature no catalyst comes off the tane. This result suggests that pentane cannot extract
alumina surface in pentane in the presence of 1- 6 from 6/Al2O3 at room temperature.
octene, and that any supported metathesis-active spe-
In order to test whether CH2=CHCH2BACTHUNGTERNN(UG pin) itself,
cies cannot be accessed by 1-octene at room tempera- aside from any metathesis reaction, can extract 6 into
ture in pentane. At 808C, 25% conversion to 47% Z- solution, we designed experiments that involve addi-
product is observed. It is not known whether the tion of (i-Pr)B
ACHTUNGTRENN(UNG pin) to 1-octene reactions. When (i-
active species at 808C is/are on the surface, in solu- Pr)B(pin) (1 equiv. relative to 1-octene) was added to
AHCTUNGTRENNUNG
tion, or both. However, most of the active species no a 1-octene reaction, the % Z product increased from
longer resemble 5, as discussed above, since Z-selec- 30% to 49% (Table 4, entries 1 and 2). These results
tivity is not high. The possibility that traces of intact 5 suggest that (i-Pr)B
are present in 5/Al2O3 and can be leached from the surface. After stirring a suspension of 6/Al2O3 in pen-
alumina surface cannot be eliminated. In contrast, 6/ tane (1 mL) containing (i-Pr)B(pin) (1 equiv. relative
ACHTUNGTNER(NUNG pin) removes 6 from the alumina
AHCTUNGTRENNUNG
Al2O3 is active for 1-octene metathesis homocoupling to 1-octene) for 16 h, the mixture was filtered to give
at room temperature, but only 30% Z is observed; Z- a light yellow filtrate and off-white solid. The filtrate
selectivity by intact 6 on alumina must be lower than gave 93% Z and 45% conversion, results which are
in solution as a consequence of what can collectively similar to the homogenous reaction employing 6
be called “surface effects,” including those that alter (93% Z, 57% conversion, Table 1). The solid gave
the nature of the transition states as a consequence of 84% Z and 44% conversion. These results suggest
Adv. Synth. Catal. 2011, 353, 1985 – 1992
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1989

Jian Yuan et al.
% Conv.
FULL PAPERS
Table 4. 1-Octene homocoupling in the presence of (i-Pr)BACHTUNTRGNEUNG(pin).
Entry
Catalyst and Additive
% Z
1
2
3
4
5
6
7
8
6/Al2O3
30
49
93
84
0
64
>98
72
47
73
45
44
0
45
3
19
6/Al2O3 +(i-Pr)B
6/Al2O3 +(i-Pr)B
6/Al2O3 +(i-Pr)B
5/Al2O3
5/Al2O3 +(i-Pr)B
5/Al2O3 +(i-Pr)B
5/Al2O3 +(i-Pr)B
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
G
AHCTUNGTRENNUNG
A
that (i-Pr)BACHTUNGTRENNUNG(pin) is able to remove a significant sulting altered catalyst is not readily removed from
amount of 6 from the alumina surface. However, we the surface by polar solvents or Lewis basic ether
cannot exclude the possibility that (i-Pr)B(pin) ad- functionalities. Since 5’ behaves like 5, the difference
AHCTUNGTRENNUNG
sorbed on the alumina surface can also help increase between 5 and 6 can be traced to the difference be-
the selectivity and conversion of 6 bound to the sur- tween the dimethylamino group being present or not.
face.
Perhaps the most surprising results are that 5/Al2O3 coupling by ethers or RB
will convert (allyl)B(pin) to a 78% Z product (14% These studies suggest that a Lewis basic site on an
conversion; Table 1) and that 1-octene is homocou- isolated homogeneous catalyst is useful for supporting
pled by 5/Al2O3 in the presence of (i-Pr)B(pin) that catalyst on g-alumina in a non-destructive
(Table 4; runs 6 and 8), but not in the absence of (i- manner, that catalysts are likely to remain on the alu-
Pr)B(pin) (Table 4; run 5). Run 7 in Table 4 suggests mina surface in hydrocarbon solvents during metathe-
5/Al2O3 appears to be activated toward Z-selective
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
that the observed activity and selectivity of 5/Al2O3 is sis reactions with hydrocarbon substrates, and that
not a consequence of traces of 5 being released into even catalysts that do not contain a Lewis basic site
solution. One explanation is that RB
ACHTUNGTRENNUNG
assist homocoupling of (allyl)B(pin) or (i-Pr)BCAHTUNGTRENNUNG
U
5/Al2O3 on the alumina surface, not in solution. It
should be noted that even the presence of ether itself
(Table 3, run 7–9) leads to 47–57% Z homocoupling
product of 1-octene. At this stage we cannot explain
exactly how homocoupling is accomplished to give
product with a % Z of 50 or above with what we
have proposed is altered 5 on alumina.
Experimental Section
General Information
Metathesis homocoupling reactions of allyl benzyl All manipulations were conducted under a nitrogen atmos-
phere in a Vacuum Atmospheres glovebox or using Schlenk
techniques. All glassware was oven-dried prior to use. Ether,
pentane, dichloromethane, toluene, and benzene were de-
gassed with dinitrogen and passed through activated alumi-
na columns. All dried and deoxygenated solvents were
stored over molecular sieves in a nitrogen or argon-filled
ether roughly parallel those involving (allyl)BACTHNUGTRNEUNG(pin),
although conversions are lower (Table 1). The high Z-
selectivities, but low conversions, can be explained if
traces of unaltered 5 are removed from 5/Al2O3.
glovebox. NMR spectra were recorded on
a Varian
Conclusions
500 MHz spectrometer at room temperature. Chemical
1
shifts for H/13C spectra were referenced to the residual res-
onances of the deuterated solvent (1H d: C6D6, 7.16 ppm;
The Z-selectivity can be used to interrogate the
nature of MAP catalysts supported on g-alumina. The CDCl3, 7.26 ppm; 13C d: C6D6, 128.06 ppm; CDCl3,
77.16 ppm) and are reported as parts per million relative to
tetramethylsilane. The following abbreviations refer to the
multiplicity: s) singlet, d) doublet, t) triplet, m) multiplet,
br) broad signal. 1-Octene, allylboronic acid pinacol ester
and allylbenzene were dried over CaH2 and vacuum trans-
ferred. Allyl benzyl ether was dried over CaH2 and distilled.
dimethylamino group in 6/Al2O3 results in 6 binding
largely non-destructively to g-alumina. However,
there is no evidence that 6/Al2O3 is highly Z-selective
as a truly heterogeneous catalyst. Polar solvents or
Lewis basic ether functionalities can remove 6 from
6/Al2O3 to yield results similar to those observed with
6 itself in solution. In contrast, 5 is adsorbed onto alu-
mina to give 5/Al2O3 and relatively large quantities of
HIPTOH. Therefore, the majority of 5 appears to be
HO-HIPT[6]
(NAr)(C3H6)
,
W
G
E
(Pyr)
N
and
W
G
E
N
ACHTUNGTRENNUNG
to literature procedures. i-PrB(pinacolate) was synthesized
in a manner analogous to that employed for EtB(pinaco-
altered in the reaction of 5 with alumina and the re- late).[12]
1990
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1985 – 1992

Preparation of Tungsten-Based Olefin Metathesis Catalysts Supported on Alumina
147.9, 147.0, 146.1, 143.3, 134.7, 134.0, 120.7, 115.0, 97.1 (O-
Experimental Procedures and Data
CH2-O), 54.9 (MeO), 41.2 (NMe2), 34.4, 30.9, 25.9, 24.3,
23.4; HR-MS (ESI): m/z=586.4623, calcd. for C40H60NO2
[M+H]+: 586.4624.
HO-HIPT-Br (1): HO-HIPT (10.2 g, 20.4 mmol) was dis-
solved in acetic acid (200 mL). Bromine (1.57 mL,
30.7 mmol) was added at room temperature. A yellow pre-
cipitate formed in a few minutes. After 16 h, Na2SO3 solu-
tion (10%, 200 mL) was added to quench excess Br2. The
off-white solid was filtered off and washed with water
(3 mL). The solid was dissolved in Et2O (300 mL) and the
solution was washed with water and dried with MgSO4. The
MgSO4 was filtered off and the ether removed to give the
product as an off-white solid. The product was recrystallized
from Et2O/hexane to give white crystals; yield: 11.28 g
HO-HIPT-NMe2 (4): CH3OCH2O-HIPT-NMe2 (5.000 g,
8.55 mmol) was dissolved in THF (40 mL). Isopropyl alcohol
(20 mL) and HCl (30 mL) were added at room temperature.
The mixture was heated to 508C in a Schlenk bomb. A pre-
cipitate formed in a few minutes. After 12 h, the mixture
became a clear solution and the solvent was removed under
vacuum. Dichloromethane (200 mL) and water (200 mL)
were added. The organic layer was separated and washed
with water until the reached pH 5–6. The organic part was
washed with brine and dried with MgSO4. After filtration,
removal of solvent gave an off-white solid. Chromatography
purification gave a pure off-white solid product; yield:
1
(96%). H NMR (500 MHz, CDCl3, 208C): d=7.23 (s, 2H,
ArH of phenol ring), 7.10 (s, 4H, ArH of Trip), 4.57 (s, 1H,
OH),2.95 (sept, J=7 Hz, 2H, p-CHMe2), 2.70 (sept, J=
7 Hz, 4H, o-CHMe2),1.30 (d, J=7 Hz, 12H, CHMe2), 1.17
(d, J=7 Hz, 12H, CHMe2), 1.09 (d, J=7 Hz, 12H, CHMe2);
13C NMR (125 MHz, CDCl3, 208C): d=150.46, 149.43,
147.72, 132.54, 129.65, 128.83, 121.42, 112.23, 34.51, 30.92,
24.47 24.21; HR-MS (ESI): m/z=577.3040, calcd. for
C36H50OBr [M+H]+: 577.3045.
1
3.81 g (82%). H NMR (500 MHz, CDCl3, 208C): d=7.10 (s,
4H, ArH of Trip), 6.55 (s, 2H, ArH of phenol ring), 4.05 (s,
1H, HO), 2.95 (sept, J=7 Hz, 2H, p-CHMe2), 2.88 (s, 6H,
NMe2), 2.82 (sept, J=7 Hz, 4H, o-CHMe2), 1.32 (d, J=
7 Hz, 12H, CHMe2), 1.17 (d, J=7 Hz, 12H, CHMe2), 1.10
(d, J=7 Hz, 12H, CHMe2); 13C NMR (125 MHz, CDCl3,
208C): d=148.6, 147.0, 144.4, 143.3, 131.87, 121.2, 115.2,
41.7 (NMe2), 34.4, 30.8, 24.6, 24.4, 24.2; anal. calcd. for
C38H55NO: C 84.23, H 10.23, N 2.58; found: C 84.19, H 9.96,
N 2.70.
CH3OCH2O-HIPT-Br
(2):
HO-HIPT-Br
(2.000 g,
3.46 mmol) was dissolved in THF (50 mL) and NaH
(249 mg, 10.38 mmol) was added at room temperature.
After stirring the mixture for 12 h, CH3OCH2Cl (0.43 mL,
6.92 mmol) was added at room temperature. After 2 h,
water (20 mL) was added to quench the reaction. The mix-
ture was extracted with Et2O and the ether extract was
dried with MgSO4. The mixture was filtered and the solvent
was removed under vacuum to give an off-white solid. The
product was recrystallized from hexane as white crystals;
Synthesis of 5’ from 5: To a 50-mL round-bottom flask in
the glovebox were added 60 mg of 5 (0.062 mmol), 5 mL
benzene,
and
80 mL
tert-butylethylene
(10 equiv.,
0.620 mmol). This mixture was allowed to stir for 1 hour,
after which time the solvent was removed under vacuum. To
the flask were added 5 mL pentane and 80 mL tert-butylethy-
lene. This mixture was allowed to stir for 1 hour, after which
time the solvent was removed under vacuum. This process
was repeated seven more times (using pentane as solvent).
1H NMR after the nine addition cycles showed that 40% of
the original 5 had been converted into 5’ (alkylidene proton
signal at 9.557 ppm in C6D6). The remaining 60% of 5 ap-
peared to have decomposed as suggested by the presence of
free HOHIPT and the absence of the metallacycle proton
signals. It is possible that fewer than nine addition cycles
would be sufficient for full conversion, provided a high con-
centration of tert-butylethylene is used and the solution is
taken to dryness under vacuum each time.
1
yield: 2.109 g (98%). H NMR (500 MHz, CDCl3, 208C): d=
7.29 (s, 2H, ArH of phenol ring), 7.03 (s, 4H, ArH of Trip),
4.15 (s, 2H, OCH2O), 2.90 (sept, J=7 Hz, 2H, p-CHMe2),
2.67 (sept, J=7 Hz, 4H, o-CHMe2), 2.38 (s, 3H, OCH3),
1.26 (d, J=7 Hz, 12H, CHMe2), 1.18 (d, J=7 Hz, 12H,
CHMe2), 1.13 (d, J=7 Hz, 12H, CHMe2); 13C NMR
(125 MHz, CDCl3, 208C): d=151.7, 148.8, 146.9, 136.6,
133.3, 131.9, 120.9, 116.0, 97.0, 55.3, 34.4, 31.0, 25.6, 24.3
23.3: HR-MS (ESI): m/z=638.3567, calcd. for C38H57NO2Br
[M+NH4]+: 638.3573.
CH3OCH2O-HIPT-NMe2
(3):
Pd2dba3
(124.5 mg,
0.136 mmol, 4% mol) and X-Phos (129 mg, 0.271 mmol)
were dissolved in THF (4 mL) and the mixture was stirred
for 1 hour. CH3OCH2O-HIPT-Br (2.109 g, 3.39 mmol) and
sodium tert-butoxide (423.1 mg, 4.41 mmol) were suspended
in THF (10 mL) in a Schlenk bomb. The Pd solution was
added to this mixture to give a purple solution. Dimethyl-
W
G
E
N
ACHTUNGTRENNUNG(NAr)-
A
N
ACHTUNGTRENNUNG
solved in benzene (~25 mL) in a Schlenk bomb. HO-HIPT-
NMe2 (198 mg, 0.366 mmol) was added at room tempera-
ture. The mixture was heated at 808C for two days. The sol-
vent was removed under vacuum and the residue was ex-
tracted into pentane. The solvent was removed under
vacuum to give a yellow foam. Recrystallization of the
crude product from pentane gave a analytically pure orange
product; yield: 225 mg (56%). 1H NMR (500 MHz, C6D6,
208C): d=9.90 (s, 1H, W=CH), 7.36 (s, 2H, Ar-H), 7.24 (s,
2H, Ar-H), 7.16–7.12 (m, 4H, Ar-H), 7.01–6.92 (m, 4H, Ar-
H), 6.39 (s, 2H, Ar-H), 6.10 (s, 2H, Pyr-H), 5.91 (s, 2H, Pyr-
H), 3.13 (sept, 2H, J=7 Hz, CHMe2), 3.07 (sept, 2H, J=
7 Hz, CHMe2), 2.89 (sept, 2H, J=7 Hz, CHMe2), 2.85 (sept,
2H, J=7 Hz, CHMe2), 2.36 (s, 6H, NMe2), 1.31(d, 6H, J=
7 Hz, CHMe2), 1.28 (d, 6H, J=7 Hz, CHMe2), 1.20–1.12 (m,
30H), 1.07 (d, 12H, J=7 Hz, CHMe2); 13C NMR (125 MHz,
ACHTUNGTRENNUNGamine (2.04 mL, 2M, 4.07 mmol) was added and the mixture
was heated to 808C. A white precipitate formed after a few
minutes. After 1 day, the solution was cooled to room tem-
perature and filtered through a silica gel and washed with
CH3COOC2H5 to give an orange solution. All volatiles were
removed from the filtrate to yield a light orange oil product.
Chromatographic separation gave a pure off-white solid
1
product; yield: 1.442 g (72%). H NMR (500 MHz, CDCl3,
208C): d=7.04 (s, 4H, ArH of Trip), 6.54 (s, 2H, ArH of
phenol ring), 4.11 (s, 2H, OCH2O), 2.92 (sept, J=7 Hz, 2H,
p-CHMe2), 2.91 (s, 6H, NMe2), 2.82 (sept, J=7 Hz, 4H, o-
CHMe2), 2.38 (s, 3H, OCH3), 1.28 (d, J=7 Hz, 12H,
CHMe2), 1.20 (d, J=7 Hz, 12H, CHMe2), 1.16 (d, J=7 Hz,
12H, CHMe2); 13C NMR (125 MHz, CDCl3, 208C): d=
Adv. Synth. Catal. 2011, 353, 1985 – 1992
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1991

Jian Yuan et al.
FULL PAPERS
C6D6, 208C): d=262.9 (Mo=C),152.2, 151.1, 150.7, 148.5,
147.6, 147.3, 145.8, 134.7, 131.9, 128.4, 126.3, 126.1, 125.0,
122.9, 122.4, 121.7, 120.9, 115.4, 52.4, 34.7, 33.5 (br), 32.0
(br), 31.0 (br), 28.1 (br), 26.2, 25.6, 24.6, 24.3, 23.8, 23.2;
anal. calcd. for C64H87N3OW: C 69.99, H 7.98, N 3.83; found:
C 69.85, H 7.91, N 3.59.
sis (CENTC). We also thank the National Science Founda-
tion for departmental X-ray diffraction instrumentation
(CHE-0946721).
W
G
ACHTUNGTRENNUNG(Pyr)ACHTUNGTRENNUNG(C3H6)(O-HIPT-NMe2) (7a): WAHCTUNGERTG(NNUN NAr)ACHTNGUTREN(NUGN Pyr)-
ACHTUNGTRENNUNG
References
dissolved in C6D6 (0.5 mL) in a J-Young tube. The solution
was subjected to three freeze/pump/thaw cycles and ethyl-
ene (1 atm) was then added to the solution at room temper-
ature. After 30 min, the solution became reddish orange. A
proton NMR spectrum showed that all of the starting mate-
rial had been converted to product: 1H NMR (500 MHz,
C6D6, 208C): d=7.43 (s, 2H, Pyr-H), 7.18 (s, 4H, Trip-H),
6.92 (d, 2H, J=8 Hz, NAr-H), 6.76 (t, 1H, J=8 Hz, NAr-
H), 6.62 (s, 2H, Ar), 6.15 (s, 2H, Pyr-H), 4.21 (brm, 2H,
WCHa), 3.62 (sept, 2H, J=7 Hz, CHMe2), 3.55 (brm, 2H,
WCHa), 3.12(sept, 2H, J=7 Hz, CHMe2), 2.87 (m, 4H,
CHMe2), 2.51 (s, 6H, NMe2), 1.40 (br, 6H, CHMe2), 1.29 (d,
24H, J=7 Hz, CHMe2), 1.25 (br, 6H, CHMe2), 1.00 (d,
12H, J=7 Hz, CHMe2), À0.80 (br, 1H, WCHb), À1.12 (br,
1H, WCHb); 13C NMR (125 MHz, C6D6, 208C): d=151.7,
148.8, 148.3, 147.5, 147.3, 136.5, 131.6, 131.1, 128.4, 127.3,
127.0, 126.4, 126.1, 122.9, 121.3, 120.4 (br), 116.7, 110.0, 97.4
(br, WCa), 35.0, 31.7 (br), 30.9 (br), 28.4 (br), 26.6 (br),
24.5(br), 23.1 (br), À3.64 (br, WCb).
[1] a) R. R. Schrock, Chem. Rev. 2002, 102, 145; b) R. R.
Schrock, Chem. Rev. 2009, 109, 3211.
[2] a) Handbook of Metathesis, (Ed.: R. H. Grubbs),
Wiley-VCH, Weinheim, 2003, Vols. 1–3; b) G. C. Vou-
gioukalakis, R. H. Grubbs, Chem. Rev. 2010, 110, 1746.
[3] M. R. Buchmeiser, Chem. Rev. 2009, 109, 303.
[4] a) P. Preishuber-Pflꢀgl, P. Buchacher, P. E. Eder, R. M.
Schitter, R. Stelzer, J. Mol. Catal. A 1998, 133, 151;
b) S. I. Wolke, R. Buffon, J. Mol. Catal. A 2000, 160,
ˇ
ˇ
181; c) H. Balcar, N. Zilkovꢃ, J. Sedlꢃcek, J. Zednꢄk, J.
Mol. Catal. A 2005, 232, 53; d) X. Solans-Monfort, C.
Copꢅret, O. Eisenstein, J. Am. Chem. Soc. 2010, 132,
7750.
[5] a) Z. Huang, E. Rolfe, E. C. Carson, M. Brookhart,
A. S. Goldman, S. H. El-Khalafy, A. H. R. MacArthur,
Adv. Synth. Catal. 2010, 352, 125; b) Z. Huang, M.
Brookhart, A. S. Goldman, S. Kundu, A. Ray, S. L.
Scott, B. C. Vicente, Adv. Synth. Catal. 2009, 351, 188.
[6] C. Stanciu, M. M. Olmstead, A. D. Phillips, M. Stender,
P. P. Power, Eur. J. Inorg. Chem. 2003, 3495.
[7] a) M. M. Flook, A. J. Jiang, R. R. Schrock, P. Mꢀller,
A. H. Hoveyda, J. Am. Chem. Soc. 2009, 131, 7962;
b) A. J. King, Y. Zhao, R. R. Schrock, A. H. Hoveyda,
J. Am. Chem. Soc. 2009, 131, 16630.
[8] a) M. M. Flook, L. C. H. Gerber, G. T. Debelouchina,
R. R. Schrock, Macromolecules 2010, 43, 7515; b) S. J.
Malcolmson, S. J. Meek, E. S. Sattely, R. R. Schrock,
A. H. Hoveyda, Nature 2008, 456, 933; c) S. J. Meek,
S. J. Malcolmson, B. Li, R. R. Schrock, A. H. Hoveyda,
J. Am. Chem. Soc. 2009, 131, 16407; d) R. R. Schrock,
A. J. King, S. C. Marinescu, J. H. Simpson, P. Mꢀller,
Organometallics 2010, 29, 5241; e) S. C. Marinescu,
R. R. Schrock, P. Mꢀller, A. H. Hoveyda, J. Am. Chem.
Soc. 2009, 131, 10840.
W
G
ACHTUNGTRENNUNG
W
N
N
ACHTUNGTRENNUNG
0.073 mmol) was dissolved in pentane (3 mL) in a Schlenk
bomb. The solution was subjected to three freeze/pump/
thaw cycles and ethylene (1 atm) was then added to the so-
lution at room temperature. The sample was stirred for
30 min to give a yellow-brown solution. The solvent was re-
moved to give a yellow-brown oil, which was extracted with
pentane. The mixture was filtered through Celite and the ex-
tract was concentrated to 1 mL and left at À308C overnight.
Filtration yielded the analytically pure orange product;
yield: 35.5 mg (41%). The filtrate was stored at À308C in
order to obtain X-ray quality crystals. In C6D6 7b quickly
dissolved to yield
WACHTUNGTREGUN(NN NAr)ACHTNUTRGEGN(NNU Pyr)ACHTUNGTRENN(UNG CH2)(O-HIPT-NMe2),
CH2=CHCMe2Ph, 6, and 7a. Most of the NMR resonances
overlapped with each other, making it difficult to assign the
peaks of 7b. Anal. calcd. for C66H91N3OW: C 70.38, H 8.14,
N 3.73; found: C 70.23, H 8.24, N 3.81.
CCDC 830541 contains the supplementary crystallograph-
ic data for compounds 6 and 7b. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[9] T. Kreickmann, S. Arndt, R. R. Schrock, P. Mꢀller, Or-
ganometallics 2007, 26, 5702.
[10] A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn, G. C.
Verschoor, J. Chem. Soc. Dalton Trans. 1984, 1349.
[11] a) J. Feldman, J. S. Murdzek, W. M. Davis, R. R.
Schrock, Organometallics 1989, 8, 2260; b) J. Feldman,
W. M. Davis, J. K. Thomas, R. R. Schrock, Organome-
tallics 1990, 9, 2535; c) J. Feldman, R. R. Schrock, Prog.
Inorg. Chem. 1991, 39, 1.
[12] J. Stymiest, G. Dutheuil, A. Mahmood, V. K. Aggarwal,
Angew. Chem. 2007, 119, 7635; Angew. Chem. Int. Ed.
2007, 46, 7491.
Acknowledgements
This work was supported by the NSF (CHE-0650456) as part
of the Center for Enabling New Technologies through Cataly-
1992
asc.wiley-vch.de
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 1985 – 1992