Tandem couplings of imines and other unsaturated organic compounds with a half-open titanocene

Article abstract of DOI:10.1039/a901304k

The half-open titanocene Ti(C₅H₅)(2,4-C₇H₁₁)(PEt₃) (C₇H₁₁ = dimethylpentadienyl) has been found to incorporate a second equivalent of PhC(H)=NMe more slowly than the first, with coupling occurring between the imine's carbon atom and the dienyl's terminal CH₂ groups. The structure of this complex has been confirmed through a diffraction study. The fact that a second imine could be incorporated has allowed mixed couplings involving initial incorporation of one equivalent of imine followed by another type of unsaturated compound. Thus, after addition and coupling of one equivalent of PhC(H)-NPrⁱ, addition of an excess of acetone or Bu^tNC leads to the respective incorporation of one or two equivalents of these species. X-Ray diffraction studies have been carried out for each, and reveal that one equivalent of ketone or nitrile has undergone coupling. A reaction involving the same imine and p-MeC₆H₄NC in the presence of Bu^tOH was found to lead to the incorporation of three equivalents of the isocyanide along with one equivalent of the alcohol. The nature of the coupled product has been revealed through a diffraction study, and a reasonable mechanism proposed for the product's formation.

Full text of DOI:10.1039/a901304k

View Article Online / Journal Homepage / Table of Contents for this issue
Tandem couplings of imines and other unsaturated organic
compounds with a half-open titanocene
Robert Tomaszewski, Atta M. Arif and Richard D. Ernst*
Department of Chemistry, University of Utah, Salt Lake City, Utah 84112, USA
Received 16th February 1999, Accepted 23rd March 1999
The half-open titanocene Ti(C5H5)(2,4-C7H11)(PEt3) (C7H11 = dimethylpentadienyl) has been found to incorporate
a second equivalent of PhC(H)᎐NMe more slowly than the first, with coupling occurring between the imine’s carbon
᎐
atom and the dienyl’s terminal CH2 groups. The structure of this complex has been confirmed through a diffraction
study. The fact that a second imine could be incorporated has allowed mixed couplings involving initial incorporation
of one equivalent of imine followed by another type of unsaturated compound. Thus, after addition and coupling of
one equivalent of PhC(H)᎐NPri, addition of an excess of acetone or ButNC leads to the respective incorporation of
one or two equivalents of these species. X-Ray diffraction studies have been carried out for each, and reveal that one
equivalent of ketone or nitrile has undergone coupling. A reaction involving the same imine and p-MeC6H4NC in
the presence of ButOH was found to lead to the incorporation of three equivalents of the isocyanide along with one
equivalent of the alcohol. The nature of the coupled product has been revealed through a diffraction study, and a
reasonable mechanism proposed for the product’s formation.
Introduction
Half open titanocenes such as Ti(C5H5)(2,4-C7H11)(PEt3) con-
tain two delocalized 5 electron donor ligands, one open and one
R
Ti
N
closed, the former of which is both more strongly bound and
more reactive than the latter.1,2 To date, the most common reac-
tions such species have been found to undergo are couplings
with unsaturated organic molecules, such as alkynes, nitriles,1
isocyanides,2 alkynes,3 imines,2 and ketones.2 In these reactions
the pentadienyl ligand may commonly engage in as many as
three couplings,† thereby serving in a formal sense as a penta-
dienyl trianion synthon [or a tris(Grignard) reagent]. In the
cases in which polyfunctionalization has been observed, as with
ketones and alkynes, one could generally not isolate intermedi-
ate products arising from fewer couplings, which appeared to be
a significant limitation to this chemistry. However, it has now
been observed that previously reported mono(imine) coupling
products can incorporate a second imine, thus opening the door
to a variety of mixed coupling products involving at least one
imine. Here we report on the mixed species in which the second
coupling partner can be a ketone, isocyanide, or another imine.
Ti
PEt3
C6H5
H
2a, R = CH3; b, R = i-C3H7
1
observed for the imine’s coupled carbon atom, which has now
selectively become a chiral center. While the spectroscopic data
do not unambiguously establish the configuration at that atom,
structural determinations of the mixed coupling products (see
below) all reveal a configuration identical to that shown above,
with the phenyl substituent directed away from the open edge
of the pentadienyl fragment. No evidence for the opposite
configuration was obtained for any of the isolated products,
although it is possible that smaller amounts of the alternative
products could have been produced in these reactions.
Results and discussion
The reaction of Ti(C5H5) (2,4-C7H11)(PEt3) 1 with one equiv-
alent of C H CH᎐NCH had earlier been reported to yield an
᎐
6
5
3
Ti
N
N
oil, whose spectroscopic characterization suggested structure
2a,2 which is consistent with a variety of other reaction prod-
ucts derived from organometallic compounds and imines.4 It
C6H5
H5C6
H
H
has now been found that the related imine PhCH᎐NPri under-
᎐
3
goes a similar reaction, yielding a crystalline solid product, 2b.
The occurrence of a single coupling process at one end of the
pentadienyl ligand is readily evidenced by the observation of
a small J(13C–H) value for one of the terminal CH2 groups of
the pentadienyl fragment (128 vs. 150 Hz), consistent with sp3
hybridization. A similarly small J(13C–H) value of 132 Hz is
For complex 2a, but not 2b, prolonged exposure in solution
to an excess of the imine led to slow conversion from deep red
into bright green, reflecting the incorporation and coupling of a
second equivalent of the imine. Compared to 2a and 2b, the 1H
and 13C NMR spectra of the product 3 are much simplified,
signifying the reestablishment of formal mirror plane sym-
metry in the molecule. Thus, a single resonance is observed
for the equivalent CH2 groups in the 1 and 5 positions of
the original pentadienyl fragment, with a J(13C–H) value of
† Once a third formal insertion between a Ti–C bond occurs the
titanium center has formally reached the ϩ4 oxidation state. Additional
couplings may still occur,2 but obviously do not lead to any further
oxidation of the titanium center.
J. Chem. Soc., Dalton Trans., 1999, 1883–1890
1883

View Article Online
Table 1 Pertinent bonding parameters (distances in Å, angles in Њ) for
compound 3
Ti–N1
Ti–N2
Ti–C2
Ti–C3
Ti–C4
Ti–C24
Ti–C25
Ti–C26
Ti–C27
Ti–C28
N1–C8
N1–C15
1.959(3)
1.955(3)
2.252(4)
2.354(4)
2.575(4)
2.389(4)
2.443(5)
2.456(5)
2.372(5)
2.334(5)
1.479(5)
1.461(5)
C1–C2
C1–C8
C2–C3
C2–C6
C3–C4
C4–C5
C4–C7
C5–C16
C8–C9
C16–C17
N2–C16
N2–C23
1.513(6)
1.521(6)
1.421(6)
1.505(6)
1.397(6)
1.507(6)
1.508(6)
1.519(6)
1.529(6)
1.534(6)
1.468(5)
1.442(5)
N1–Ti–N2
Ti–N1–C8
Ti–N1–C15
C8–N1–C15
N1–C8–C1
N1–C8–C9
C9–C8–C1
C8–C1–C2
C1–C2–C3
C1–C2–C6
C3–C2–C6
97.0(1)
116.3(2)
130.4(3)
111.5(3)
106.5(3)
115.1(3)
112.3(4)
113.7(4)
125.3(4)
114.9(4)
114.9(4)
C2–C3–C4
C3–C4–C5
C3–C4–C7
C5–C4–C7
C4–C5–C16
C5–C16–C17
C5–C16–N2
C17–C16–N2
C16–N2–Ti
C16–N2–C23
C23–N2–Ti
133.6(4)
126.8(4)
117.7(4)
113.4(4)
113.1(4)
113.2(3)
106.4(3)
114.7(4)
129.7(3)
111.3(3)
119.0(3)
Fig. 2 Perspective view of compound 4.
for their C–N–C angles). Thus, the Ti–N1–C8 angle of
116.3(2)Њ is less than the Ti–N2–C16 angle of 129.7(3)Њ, while
the opposite is true for the Ti–N(1 or 2)–C(15 or 23) angles,
130.4(3) vs. 119.0(3)Њ. Whether this is a result of an attempt to
relieve eclipsing CH3 ؒ ؒ ؒ CH3 interactions, to optimize the two
π interactions with the titanium center, or simply is a response
to packing forces is unclear. In any event, the upward drift of
N1 relative to N2 then seems responsible for the asymmetry in
the titanium–allyl bonding, as this leads to more of a trans
relationship with C4.
Ti
N
O
H5C6
H
Fig. 1 Solid state structure of compound 3.
4
127 Hz, indicative of sp3 hybridization. Once again, a similar
J(13C–H) value (135 Hz) is observed for the imine’s coupled
carbon atoms.
As one would expect from the fact that two equivalents of a
ketone can couple to a half-open titanocene, addition of acet-
one to compound 2b leads to the anticipated coupling between
the carbonyl group and the previously uncoupled dienyl
terminus. This is quite evident from the J(13C–H) values of
125 and 118 Hz for the two CH2 groups, reflecting formal sp3
hybridization. Thus, although 2b will not incorporate a second
imine, a smaller and more reactive ketone may be added. How-
ever, a large difference in chemical shift is observed for the C2
and C4 positions of the original dienyl fragment, δ 83.7 vs.
126.7, which should reflect a major bonding difference relative
to 3. This is in fact observed from the structural result (Fig. 2,
Table 2). As can be seen, in this case the product may be con-
sidered to be a Ti(C5H5)(σ-allyl)(alkoxide)(amide) species 4.
The σ-allyl interaction is clearly indicated by the short Ti–C2
distance of 2.153(3) Å, and the sizeable asymmetry in the C2–
C3 and C3–C4 distances [1.512(4) vs. 1.338(5) Å]. There are no
apparent overriding steric or geometric reasons for the change
in allyl co-ordination mode, but since a σ interaction has
also been found in bis(ketone) coupling products the formal
presence of potentially 5 electron donor alkoxide group(s)
would seem to be implicated. The Ti–O–C18 angle of 146.1(2)Њ,
while not the ideal 180Њ value one might expect, is nonethe-
less significantly greater than the value of ca. 133Њ expected for
3 electron donation,5 and the Ti–O distance of 1.803(2) Å is
clearly shorter than related examples which are considered as 3
electron donors,2,5 so there is good reason to implicate at least
some contribution of the alkoxide ligand as a 5 electron donor.
The Ti–N distance of 1.914(3) Å is actually slightly shorter
than the values for 3, and with the nearly planar arrangement
about N clearly reflects its formal donation of 3 electrons. It can
A single crystal diffraction study has confirmed the structural
formulation for 3 (Fig. 1, Table 1). During the coupling process,
the imine’s phenyl substituents have been oriented away from
the open edge of the pentadienyl fragment, thereby selectively
generating chiral centers at each location. Based on the general
structural result, one can consider the product formally to be a
Ti(C5H5)(allyl)(amide)2 complex. The Ti–N bond distances are
essentially equivalent at 1.957(2) Å, and when taken together
with the nearly trigonal planar arrangements about N1 and N2
this suggests that the amides are functioning as 3 electron
donors, through an additional π interaction each. These π
interactions may be exerting a significant effect on the Ti–C5H5
bonding, as the two carbon atoms (C25, C26) most opposite to
N1 and N2 are clearly furthest from Ti at ca. 2.450(4) Å, with
C24 and C27 closer at ca. 2.381(4) Å, and C28 closest at
2.334(5) Å. Also notable is the asymmetry in the titanium–allyl
bonding. With the Ti–C(2–4) distances progressively increasing
[2.252(4), 2.354(4), 2.575(4) Å], and the C2–C3 bond appearing
to be longer than C3–C4 [1.421(6) vs. 1.397(6) Å], one could
consider the possibility of some contribution of a σ-allyl com-
plex. However, in such a case a much shorter Ti–C2 distance
should be observed (see below), as well as a greater difference
between the C–C distances. It is likely best to consider the allyl
co-ordination as η3, although there may be some contribution
from a localized bonding mode (i.e., with a Ti–C2 single bond
and C3᎐C4 also co-ordinated). The origin for this asymmetry
᎐
could be a result of some asymmetry in the titanium–amide
interactions. While the angles about the C(1 vs. 5) and C(8 vs.
16) atoms are similar, those about N1 and N2 are not (except
1884
J. Chem. Soc., Dalton Trans., 1999, 1883–1890

View Article Online
Table 2 Pertinent bonding parameters (distances in Å, angles in Њ) for
compound 4
Table 3 Pertinent bonding parameters (distances in Å, angles in Њ) for
compound 5
Ti–O
Ti–N
1.803(2)
1.914(3)
2.153(3)
2.383(3)
2.457(3)
2.461(4)
2.438(3)
2.407(3)
1.470(4)
1.467(4)
O–C18
C1–C2
C1–C8
C2–C3
C2–C6
C3–C4
C4–C5
C4–C7
C5–C18
C8–C9
1.410(4)
1.547(5)
1.535(5)
1.512(4)
1.531(5)
1.338(5)
1.510(5)
1.498(5)
1.550(5)
1.521(4)
Ti–N1
Ti–N2
Ti–C6
1.973(2)
1.919(2)
2.113(2)
2.433(3)
2.422(3)
2.400(3)
2.416(3)
2.427(3)
2.217(2)
1.504(3)
1.482(3)
1.393(3)
C1–C2
C1–C7
C2–C3
C2–C6
C2–C8
C3–C4
C4–C5
C4–C9
C5–C6
N2–C19
N3–C28
N3–C29
1.538(3)
1.553(3)
1.528(3)
1.566(3)
1.542(3)
1.321(3)
1.513(3)
1.498(3)
1.540(3)
1.488(3)
1.154(3)
1.456(3)
Ti–C2
Ti–C21
Ti–C22
Ti–C23
Ti–C24
Ti–C25
N–C8
Ti–C23
Ti–C24
Ti–C25
Ti–C26
Ti–C27
Ti–C28
N1–C7
N1–C16
N2–C6
N–C15
O–Ti–N
107.9(1)
104.9(1)
85.7(1)
146.1(2)
115.0(2)
129.3(2)
C5–C4–C7
C4–C5–C18
N–C8–C1
N–C8–C9
C1–C8–C9
C8–C9–C10
C8–C9–C14
C10–C9–C14
C9–C10–C11
C10–C11–C12
C11–C12–C13
C12–C13–C14
C9–C14–C13
N–C15–C16
N–C15–C17
C16–C15–C17
O–C18–C5
116.0(3)
114.5(3)
107.7(2)
114.4(3)
111.0(3)
119.5(3)
122.4(3)
118.1(3)
120.9(3)
120.1(3)
120.0(3)
119.3(4)
121.6(3)
109.5(3)
112.4(3)
109.9(3)
107.7(2)
O–Ti–C2
N–Ti–C2
Ti–O–C18
Ti–N–C8
Ti–N–C15
N1–Ti–N2
N1–Ti–C6
N1–Ti–C28
N2–Ti–C6
N2–Ti–C28
C6–Ti–C28
Ti–N1–C7
Ti–N1–C16
C7–N1–C16
Ti–N2–C6
Ti–N2–C19
C6–N2–C19
C28–N3–C29
C2–C1–C7
C1–C2–C3
C1–C2–C6
97.92(7)
94.44(8)
94.63(8)
40.07(8)
86.80(8)
126.85(8)
112.49(12)
132.39(14)
115.0(2)
77.50(12)
148.2(2)
132.1(2)
177.3(2)
116.7(2)
114.7(2)
116.5(2)
C3–C2–C6
C2–C3–C4
C2–C3–C5
C3–C4–C9
C5–C4–C9
C4–C5–C6
Ti–C6–N2
Ti–C6–C2
Ti–C6–C5
N2–C6–C2
N2–C6–C5
C2–C6–C5
N1–C7–C1
N1–C7–C10
C1–C7–C10
Ti–C28–N3
100.3(2)
112.4(2)
110.4(2)
128.9(2)
120.7(2)
103.3(2)
62.43(11)
114.37(13)
132.2(2)
122.2(2)
119.9(2)
103.1(2)
115.4(2)
114.6(2)
113.2(2)
173.1(2)
C8–N–C15 114.6(2)
C2–C1–C8
Ti–C2–C1
Ti–C2–C3
Ti–C2–C6
C1–C2–C3
C1–C2–C6
C3–C2–C6
C2–C3–C4
C3–C4–C5
C3–C4–C7
112.7(3)
95.4(2)
106.0(2)
125.1(2)
108.9(3)
109.6(3)
110.3(3)
130.5(3)
123.4(3)
120.5(3)
be further noted that the decrease in bond distance going from
Ti–C2 to Ti–N to Ti–O [2.153(3), 1.914(3), 1.803(2) Å] is in
each case significantly greater than the decrease in covalent
radii for these non-metals,6 and also can not be fully accounted
for by electronegativity differences.7
An obvious structural difference relative to compound 3 is
the twist around the C2–C3 bond, which leads to a more sickle-
shaped dienyl fragment. Such a twist was not observed for the
bis(ketone) coupling product,2 suggesting that it may be
brought about by the imine’s isopropyl group. The alteration of
the oxygen atom location leads to significantly larger O–Ti–
(C2, N) angles relative to the C2–Ti–N angle [104.9(1) and
107.9(1) vs. 85.7(1)Њ]. A final apparent result of the altered
heteroatom locations is the slight asymmetry in Ti–C5H5 bond-
ing. As for 3, the atoms furthest from a trans location (C21,
C25) are those which are closest to the titanium center.
Fig. 3 Solid state structure of compound 5.
CN
N
more complex processes observed with aromatic isocyanides
(see below), likely the result of the additional steric bulk of the
But group.8 For a pentadienyl ligand, a simple coupling of an
isocyanide to the two dienyl ends would lead to an allyl frag-
ment, as opposed to the present alkene, and the metal would
need to remain attached to the allyl group. It is likely that the
requirement that the metal remain attached to both the putative
allyl and initial one end-coupled isocyanide prevented the other
end from coupling, at least until a second isocyanide arrived
and coupled between the first isocyanide and the other terminal
group, yielding a seven membered ring. Even after that the allyl
group may have problems interacting with the metal center, as
further incorporation and couplings of isocyanides took place.
In examining the bonding parameters for 5, one finds nothing
unusual about either the uncoupled isocyanide ligand or the Ti–
N1 interaction. In the latter case the amide seems to function as
a 3 electron donor, given a nearly planar disposition about
N1. However, the Ti–N1 distance of 1.973(2) Å is slightly to
somewhat longer than the values observed for 3 and 4. The
main point of interest is then the bonding involving the coupled
isocyanide fragment. As shown above for 5, one could consider
Ti
N
H
C6H5
CH3
5
Reactions of compound 2b with isocyanides were also investi-
gated. With ButNC, two equivalents were incorporated. In this
case spectroscopic data did not reveal much about the product
5, other than the fact that both terminal CH2 groups had
coupled [J(13C–H) = 125, 126 Hz], and that one isocyanide
apparently had not undergone any coupling (νC–N 2172 cmϪ1). A
structural determination was therefore carried out to elucidate
the mode of coupling, and pertinent bonding parameters are
presented in Table 3. As can be seen in Fig. 3, one of the iso-
cyanides has simply co-ordinated to the metal center, while the
second has had its terminal carbon atom couple to both ends of
the former diene fragment, leading to structure 5. This rather
simple mode of coupling contrasts significantly with the much
J. Chem. Soc., Dalton Trans., 1999, 1883–1890
1885

View Article Online
Table 4 Pertinent bonding parameters (distances in Å, angles in Њ) for
compound 6ؒOEt2
the Ti–C6 interaction to be a single bond, and the N2–Ti inter-
action to be that of a π amide (3 electron donation), and this
would be consistent with the short Ti–C6 distance [2.113(2) Å],
the Ti–N2 distance of 1.919(2) Å, and the nearly trigonal
disposition about N2. However, the C6–N2 distance of 1.393(3)
Å is clearly much shorter than it should be for a single bond
(cf. ca. 1.47 Å for 3 and 4). Furthermore, the angles about C6
(excluding those with Ti) average 115.1Њ, getting close to that
expected for sp2 hybridization. It may therefore be that a partial
π bond is being generated between C6 and N2, although the
orientations of their available p (or hybrid) orbitals for such an
interaction would be significantly twisted relative to the ideal
situation.9 In any case, should a net of 4 electrons be donated
from C6 and N2, the titanium center could achieve an 18
electron configuration.
Ti–O
1.760(10)
1.946(12)
1.939(11)
2.42(2)
2.45(2)
2.43(2)
2.42(2)
2.40(2)
1.45(2)
1.49(2)
1.51(2)
1.46(2)
1.43(2)
1.40(2)
N3–C32
N4–C7
N4–C14
C1–C2
C1–C16
C2–C3
C2–C6
C3–C4
C4–C5
C5–C6
C6–C7
C7–C8
C8–C9
C13–C14
1.39(2)
1.39(2)
1.39(2)
1.53(2)
1.51(2)
1.55(2)
1.61(2)
1.34(2)
1.51(2)
1.56(2)
1.51(2)
1.37(2)
1.45(2)
1.39(2)
Ti–N2
Ti–N3
Ti–C46
Ti–C47
Ti–C48
Ti–C49
Ti–C50
O–C42
N1–C16
N1–C39
N2–C6
N2–C25
N3–C8
O–Ti–N2
O–Ti–N3
101.1(5)
102.7(5)
96.0(5)
C2–C3–C4
C3–C4–C5
C3–C4–C24
C5–C4–C24
C4–C5–C6
N2–C6–C2
N2–C6–C5
N2–C6–C7
C2–C6–C5
C2–C6–C7
C5–C6–C7
N4–C7–C6
N4–C7–C8
C6–C7–C8
N3–C8–C7
N3–C8–C9
C7–C8–C9
110.6(15)
112.5(14)
125.7(17)
121.7(14)
102.7(12)
114.5(12)
109.4(11)
106.3(13)
102.3(12)
110.2(12)
114.4(13)
122.7(17)
108.1(15)
128.4(17)
124.8(17)
126.8(16)
106.9(15)
N2–Ti–N3
Ti–O–C42
C16–N1–C39
Ti–N2–C6
Ti–N2–C25
C6–N2–C25
Ti–N3–C8
Ti–N3–C32
C8–N3–C32
C7–N4–C14
C2–C1–C16
C1–C2–C3
C1–C2–C6
C3–C2–C6
169.0(11)
117.1(16)
120.8(10)
118.5(9)
120.4(12)
102.9(10)
136.4(11)
120.8(12)
109.4(14)
117.6(12)
116.8(13)
110.4(12)
99.6(12)
H
N
N
HN
Ti
O
N
6
As is often the case,10 a more complex process is followed
with aryl isocyanides. Unfortunately, some of the products
from compound 2b were not readily amenable to crystallization
and isolation. However, a very small amount of crystalline solid
was isolated in one case, as the result of a small ButOH impur-
ity in the p-tolyl isocyanide, and it was subsequently observed
that this species could be isolated in good yields by using a
stoichiometric (1:3) ratio of the alcohol and isocyanide in the
coupling reaction. Infrared and NMR data indicated that all
isocyanides had coupled, as well as the original dienyl’s two
terminal CH2 groups, but not the dienyl’s central CH group.
A diffraction study revealed the structure to be that of 6 (R =
p-MeC6H4) (Fig. 4, Table 4), in which some rather complex
couplings have occurred, along with conversion of the alcohol
into alkoxide, with transfer of the proton to a nitrogen center. A
possible mechanism for the formation of this complex is pro-
vided in Scheme 1. The key issue of contention in the mechan-
ism would be the point at which the alcohol is incorporated. It
has been observed that 2b will react with ButOH, although
relatively slowly compared to its reactions with isocyanides,
resulting in the addition of a proton to the diene ligand, yield-
ing an allyl ligand.11 Thus, it is fairly clear that 2b preferentially
incorporates an isocyanide molecule instead of an alcohol.
Given that conclusion, it seems most likely that the subsequent
isocyanide incorporations occur prior to alcohol incorporation,
although this point is not absolutely established. However, the
alcohol incorporation must occur prior to the construction of
the indole skeleton, as it is known that in the absence of the
alcohol the formation of an indole is not competitive with the
incorporation of a fourth equivalent of isocyanide.11 Hence, the
sequence presented in Scheme 1 affords a plausible explanation
for the formation of the observed product, 6. One interesting
step is that in which the coupling of an azaallyl fragment
(depicted as a π ligand, although σ co-ordination is also pos-
Fig. 4 Solid state structure of compound 6.
sible) with a p-tolyl group leads to the first fused ring precursor
to the indole. The coupling could be envisioned as a nucleo-
philic attack on the arene, or as a formal insertion of an olefin
(within the arene ring) into the Ti–C (azaallyl) bond. This step
must necessarily differ from that proposed for an indole con-
struction from a complex pentadienyl–isocyanide coupling
reaction,2‡ and thus raises the likely possibility that the fused
ring construction for the previous indole system would also
occur in a manner analogous to that proposed here.
‡ The reason a different process is proposed here derives from the fact
that in the present case, just prior to the indole construction, the com-
plex must be regarded as involving TiIV, whereas in the previous case it
could still be considered TiII.
1886
J. Chem. Soc., Dalton Trans., 1999, 1883–1890

View Article Online
C6H5
R
C6H5
H
C6H5
H
H
N
C6H5
H
N
RNC
RNC
N
RN
Ti
Ti
N
Ti
Ti
N
N
R
N
R
2b
RNC
NH
C6H5
C6H5
H
R
R
NH
C6H5
R
N
N
N
N
O
O
R
N
R
N
Ti
N
Ti
N
Ti
R
N
Me3COH
N
NH
R
NH
NH
R
R
O
C6H5
C6H5
C6H5
N
N
N
H
N
O
O
N
N
TiH
Ti
Ti
R
R
N
NH
N
R
6
(R = p-CH3C6H4)
Scheme 1
distance and the Ti–O–C angle for a variety of species;2 the
correlation roughly suggests formal 1 (sp3) and 5 (sp) electron
donation limits for the Ti–O distances to be ca. 1.86 and 1.75 Å,
respectively. Interestingly, data for some other metal systems
have been equivocal,12 perhaps in part due to sizeable steric
effects and the poorer π-donating ability of aryl oxide vs. alkox-
ide ligands.13 However, were both amides to serve as 3 electron
donors, and the alkoxide and C5H5 ligands to serve as 5 electron
donors, the complex would attain a 20 electron configuration.
There is no clear reason to believe any slippage of the C5H5
ligand is occurring, as the Ti–C distances are comparable to
those in 4; however, the Ti–N(2,3) distances average 1.943(8) Å,
appearing to be longer than the 1.914(3) Å distance in 4. It is
quite possible if not inevitable, therefore, for 6 and other species
as well, that the π donations for the amide, alkoxide, and even
Cp ligands represent only partial interactions, as metal orbitals
would not be available to accept the electrons from each ligand
orbital. In this case a parallel may be drawn with the situation
for W(PhCCPh)3(CO).14
In addition to the above coupling reactions, it was of interest
to investigate other modes of reactivity of these complexes.
In an attempt to initiate the removal of the coupled fragment
from the metal center, a reaction between 2b and C2Cl6 was
attempted. In fact, a change to bright red was observed, but the
product could only be isolated as a red oil. Nevertheless, 1H and
13C NMR spectroscopic data (see Experimental section) suggest
that the diene has been removed from the metal center, as for
example three C–H resonances experience significant downfield
shifts, which would be expected for the free diene unit in one
possible product 7. Unfortunately, crystalline material could
not be isolated even at Ϫ90 ЊC, preventing definitive character-
ization through a structural study.
Fig. 5 Variation of the Ti–O bond distances with the Ti–O–C bond
angles in related coupling products.
The structure may be considered to be pseudo-tetrahedral
with the four co-ordination sites on Ti being occupied by C5H5,
O, N2, and N3. The N–Ti–(N,O) angles are fairly similar, with
that of N2–Ti–N3 [96.0(5)Њ] being slightly smaller than the
others, perhaps due to the presence of a bridge between N2 and
N3. The angles about both N2 and N3 sum to ca. 360Њ, reflect-
ing a planar arrangement, consistent with 3 electron donation.
The Ti–O–C42 angle of 169.0(11)Њ is greater than the value of
146.1(2)Њ in 4, whose alkoxide interaction appeared to possess
some contribution as a five electron donor. In fact, the present
Ti–O distance of 1.760(10) Å is even shorter than that of
1.803(2) Å in 4, again perhaps implicating some contribution
as a five electron donor. As can be seen in Fig. 5, there is an
excellent correlation (R2 = 0.978, with the average deviation in
distance from the plot being 0.003 Å, or 1σ) between Ti–O
J. Chem. Soc., Dalton Trans., 1999, 1883–1890
1887

View Article Online
cedures. Spectroscopic data were obtained as previously
described.18 The 13C NMR spectra were not precisely inte-
grated, but numbers of carbon atoms are given in accord with
their assignments. Elemental analyses were obtained from
E ϩ R Microanalytical Labs, Robertson Microanalytical Labs,
or Desert Analytics.
Ti
Cl
Cl
N
Preparations
Ti(C H )(NPriCHPhCH CMe᎐CHCMe᎐CH ) 2b. To
a
᎐
᎐
5
5
2
2
stirred solution of Ti(C5H5)(2,4-C7H11)(PEt3) (0.50 g, 1.4
mmol) in 50 mL THF under nitrogen at Ϫ78 ЊC was added
PhC(H)NPri (0.42 mL, 2.8 mmol). After warming to room
temperature with stirring overnight, a dark red color appeared.
The solvent was then removed in vacuo, leaving behind a red
oily residue which was in turn extracted with three 50 mL por-
tions of pentane. These extracts were filtered through a Celite
pad on a coarse frit. The dark red filtrate was concentrated in
vacuo to ca. 20 mL and placed in a Ϫ90 ЊC freezer for 2 d.
Removal of the supernatant via syringe and drying in vacuo
gave 0.40 g (74%) of the product as an air-sensitive, dark red
solid (mp 81–82 ЊC) (Found: C, 74.51; H, 8.08; N, 3.89. Calc.
for C22H29NTi: C, 74.36; H, 8.08; N, 3.89%). 1H NMR
(benzene-d6, ambient): δ 7.5–7.4 (m, 5 H, Ph), 5.67 (s, 5 H, Cp),
4.50 (d, 1 H, J = 6, H-5exo), 3.65 (s, 1 H, H-3), 3.14 [septet,
J = 6.6, NCH(CH3)2], 2.98 (dd, 1 H, J = 6, 1.4, H-1), 2.67 (dd,
1 H, J = 6, 1.5, H-5endo), 2.52 (dd, 1 H, J = 14, 7, CHPh), 2.24 (d,
1 H, J = 14, H-1Ј), 1.58 (s, 3 H, CH C᎐), 1.56 (s, 3 H, CH C᎐),
7
As ButOH was found to react selectively with an isocyanide
coupling product, yielding 6, a similar reaction was attempted
with 2b. Indeed, a slow color change revealed that a reaction
had occurred, but the product could only be isolated as an oil,
precluding definitive structural characterization. Nonetheless,
relatively clean H and 13C NMR spectra were obtained. Par-
1
ticularly diagnostic was the presence of eight methyl groups,
providing a strong indication that protonation did not occur on
nitrogen (as happened for 6) but instead for the uncoupled ter-
minal CH2 group of the original dienyl fragment, presumably
yielding 8. However, the product can be expected to be a σ,
rather than π, allyl, based upon the precedents demonstrated
for other mixed allyl/alkoxide complexes2 (see below). Support
for this assignment can be obtained from the 13C NMR reson-
ances for the allylic fragment. For the mixed imine/ketone coup-
ling product 4 these were observed at δ 83.7, 139.6, and 126.7,
nearly identical to those of the proposed 8, δ 83.6, 144.5, and
121.4.
᎐
᎐
3
3
0.86 [d, 3 H, J = 6.5, CH(CH3)2] and 0.65 (d, 3 H, J = 6.6,
CH(CH3)2]. 13C NMR (benzene-d6, ambient): δ 151.7 (s, 1 C),
128–126 (5 C), 115.9 (d, 1 C, J = 154, C-3), 108.5 (d of quintets,
5 C, J = 172, 7, Cp), 95.9 (s, 1 C, C-2 or 4), 67.6 (d, 1 C, J = 132,
CHPh), 61.1 (t, 1 C, J = 150, C-5), 52.9 [d, 1 C, J = 131,
CH(CH3)2], 43.9 (t, 1 C, J = 128, C-1), 32.2 (q, 1 C, J = 132,
C6H5
CH C᎐), 30.4 (q, 1 C, J = 132, CH C᎐), 24.4 [qt, 1 C, J = 126, 4,
᎐
᎐
3
3
NCH(CH3)2] and 23.6 [qt, 1 C, J = 126, 5, NCH(CH3)2].
N
ؒؒ ؒؒ
Ti
Ti(C5H5)(NMeCHPhCH2CMe᎐CH᎐CMeCH2CHPhNMe)
3. To a stirred solution of Ti(C5H5)(2,4-C7H11)(PMe3) (1.00 g,
3.52 mmol) in 50 mL THF under nitrogen at Ϫ78 ЊC was added
PhC(H)NMe (1.00 mL, 8.11 mmol, 2.2 equivalents). Upon
warming with stirring to room temperature overnight a dark
green color appeared. The solvent was then removed in vacuo
leaving a green oily residue which was in turn extracted with
three 50 mL portions of pentane. The extracts were filtered
through a Celite pad on a coarse frit. The emerald green filtrate
was then concentrated in vacuo to ca. 20 mL and placed in a
Ϫ20 ЊC freezer for 2 d. Removal of the supernatant via syringe
and drying in vacuo gave 0.74 g (45%) of the product as an
air-sensitive dark green solid (mp 118–120 ЊC). The C5H4Me
analogue was also prepared in the same way in 40% yield
(Found: C, 75.08; H, 7.74; N, 6.12. Calc. for C28H34N2Ti: C,
O
8
The fact that a second equivalent of an imine will couple to a
pentadienyl ligand much more slowly than the first, presumably
as a result of steric influences, has allowed for the tandem coup-
lings of a pentadienyl ligand with an imine and either a ketone,
nitrile,11 other imine, or even isocyanide. In each case the two
heteroatoms become joined by a chain of seven carbon atoms.
Conceivably, the use of bulkier ketones such as acetophenone
could also lead to the selective isolation of mono(coupling)
products, thereby allowing for the subsequent incorporation
of a second coupling partner. In any event, other tandem coup-
lings involving Zr(C5H5)2(diene) complexes have also been
achieved,15 and in the cases involving heteroatom-containing
multiple bonds the heteroatoms become joined by a chain of six
carbon atoms. Certainly one can expect that, with the appropri-
ate choice of metal π complex, similar tandem couplings will be
able to lead to related species with other chain lengths separat-
ing the heteroatoms. Such possibilities are currently under
investigation.
1
75.32; H, 7.68; N, 6.27%). H NMR (benzene-d6, ambient):
δ 7.3–7.1 (m, 10 H, Ph), 5.83 (s, 5 H, Cp), 4.85 (d, 2H, J = 8,
H-1,5), 4.07 (s, 1 H, H-3), 3.60 (dd, 2 H, J = 14, 8, CHPh), 3.01
(s, 6 H, NCH3), 2.15 (d, 2 H, J = 14, H-1,5Ј) and 1.68 (s, 6 H,
CH C᎐). 13C NMR (benzene-d6, ambient): δ 147.6 (s, 2 C, Ph),
᎐
3
128.5–127.7 (m, 10 C, Ph), 116.7 (s, 2 C, C-2,4), 111.1 (d of
quintets, 5 C, J = 171, 6, Cp), 109.1 (d, 1 C, J = 170, C-3), 82.2
(d, 1 C, J = 135, CHPh), 44.1 (q, 2 C, J = 130, NCH3), 40.5 (t,
2 C, J = 127, C-1,5) and 32.2 (q, 2 C, J = 127 Hz, CH C᎐). Mass
᎐
3
spectrum (EI, 25 eV): m/z (relative intensity) 327 (9), 218 (9),
217 (61), 216 (77), 215 (10), 214 (23), 208 (31), 207 (10), 206
(43), 198 (15), 184 (20), 160 (19), 121 (35), 120 (100), 119 (34),
118 (60), 104 (9), 91 (14), 77 (8) and 42 (21).
Experimental
All operations were carried out under a nitrogen atmosphere
using either Schlenk techniques or a glove-box. Hydrocarbon,
aromatic, diethyl ether and THF solvents were distilled
under nitrogen from sodium–benzophenone. Aryl isocyanides16
and Ti(C5H5) (2,4-C7H11)(PR3) (C7H11 = dimethylpentadienyl,
R = Me or Et)1,17 were prepared according to published pro-
Ti(C H )(NPriCHPhCH CMeCH᎐CMeCH CMe O) 4. To
᎐
5
5
2
2
2
a stirred solution of compound 2b (0.92 g, 2.6 mmol) in 50 mL
diethyl ether under nitrogen at Ϫ78 ЊC was added acetone (0.23
mL, 3.1 mmol, 1.1 equivalents). The mixture was warmed with
1888
J. Chem. Soc., Dalton Trans., 1999, 1883–1890

View Article Online
stirring to 0 ЊC, changing from dark to light red. At 0 ЊC, the
solvent was removed in vacuo, leaving a red solid, which was in
turn extracted with three 50 mL portions of pentane. The
extracts were filtered through a Celite pad on a coarse frit. The
bright red filtrate was concentrated in vacuo to ca. 20 mL.
Placement of the filtrate in a Ϫ90 ЊC freezer for 2 d gave a red
solid. The supernatant was removed via syringe, and the solid
dried in vacuo to give 0.42 g (41%) of the product as an air-
sensitive bright red solid (mp 112–115 ЊC) (Found: C, 72.43; H,
8.63; N, 3.29. Calc. for C25H35NOTi: C, 72.63; H, 8.53; N,
3:1 mixture of p-tolyl isocyanide–tert-butyl alcohol (0.61 mL,
4.2 mmol:0.13 mL, 1.4 mmol). A rapid change from dark to
bright red occurred upon the addition. The mixture was
warmed with stirring to room temperature and then stirred at
room temperature for an additional hour. Next the solvent was
removed in vacuo leaving a dark red sticky solid. This was
extracted with three 50 mL portions of ether–pentane (50:50).
Then the extracts were filtered through a Celite pad on a coarse
frit. The bright red filtrate was concentrated in vacuo to ca. 20
mL, placed into a Ϫ90 ЊC freezer for 2 d, then transferred to a
Ϫ30 ЊC freezer for 3 d during which dark red crystals emerged.
Removal of the supernatant via syringe and drying of the crys-
tals in vacuo yielded 0.67 g (61%) of the product as air-stable
dark red cube-shaped crystals (mp 129–133 ЊC) (Found: C,
76.58; H, 8.01; N, 6.93. Calc. for C50H60N4OTi: C, 76.82; H,
1
3.39%.) H NMR (benzene-d6, ambient): δ 7.33–7.10 (m, 5 H,
Ph), 6.16 (s, 1 H, H-3), 6.13 (s, 5 H, Cp), 4.39 (dd, 1 H, J = 8,
12), 2.63 [septet, 1 H, J = 6.4, CH(CH3)2], 2.31 (dd, 1 H, J = 12,
15), 2.11 (dd, 1 H, J = 8, 15), 1.98 (d, 1 H, J = 12, H-1), 1.98 (s,
3 H, CH C᎐), 1.33 [2 s, 6 H, (CH ) CO], 1.30 (d, 1 H, J = 12,
᎐
3
3 2
1
H-1Ј), 1.18 (s, 3 H, CH C᎐), 1.17 [d, 3 H, J = 6.1, CH(CH ) ]
7.74; N, 7.17%). H NMR (benzene-d6, ambient; some reson-
᎐
3
3 2
and 0.91 [d, 3 H, J = 6.9 Hz, CH(CH3)2]. 13C NMR (benzene-d6,
ambient): δ 146.5 (s, 1 C, Ph), 139.6 (d, 1 C, J = 146, C-3),
126.7–129.9 (m, 4 C, Ph), 126.7 (s, 1 C, C-4), 112.6 (d of quin-
tets, 5 C, J = 171, 6, Cp), 90.3 (s, 1 C, C–O), 83.7 (s, 1 C, C-2),
61.6 (d, 1 C, J = 136, CHPh), 53.4 [d, 1 C, J = 135, CH(CH3)2],
50.5 (t, 1 C, J = 124, C-1 or 5), 48.1 (tq, 1 C, J = 128, 6, C-1 or
ances obscured): δ 7.34 (s, 1 H), 7.30–6.96 (Ph), 5.77 (s, 5 H,
Cp), 5.62 (s, 1 H, H-3), 3.81 (dd, 1 H, J = 6, 16), 2.96 (dd, 2 H,
J = 7, 15), 2.51 [septet, 2 H, J = 3.3, CH(CH3)2], 2.45 (d,
1 H, J = 16), 2.31 (d, 1 H, J ≈ 16), 2.28 (s, 3 H, CH3), 2.22 (s,
3 H, CH3), 2.16 (s, 3 H, CH3), 2.02 (dd, 1 H, J = 6, 15), 1.27 (s,
3 H, CH C᎐), 1.17 [s, 9 H, C(CH ) ], 0.99 [d, 3 H, J = 6.0,
᎐
3
3 3
5), 31.5 (q, 1 C, J = 124, CH C᎐), 30.7 (q, 1 C, J = 128, CH C᎐),
CH(CH3)2], 0.82 [d, 3 H, J = 6.3 Hz, CH(CH3)2] and 0.65 (s,
᎐
᎐
3
3
29.3 (q, 1 C, J = 123, CH3CO), 27.8 (q, 1 C, J = 125, CH3CO),
24.9 [q, 1 C, J = 125, CH(CH3)2] and 20.2 [q, 1 C, J = 124 Hz,
CH(CH3)2]. Mass spectrum (EI, 33 eV): m/z (relative intensity)
413 (16), 356 (11), 355 (36), 266 (30), 210 (14), 209 (41), 208
(100), 207 (43), 206 (91), 205 (15), 204 (14), 169 (17), 148 (56),
147 (32), 132 (61), 106 (9), 105 (16), 104 (13) and 91 (11).
3 H, CH C᎐). 13C NMR (benzene-d6, ambient): δ 150.5 (s, 1 C),
᎐
3
150.4 (s, 1 C), 147.0 (s, 1 C), 135.0–121.5 (22 C), 118.6 (d 1 C,
J = 153), 113.3 (s, 1 C), 111.8 (d, 1 C, J = 157, C-3), 109.7 (d of
quintets, 5 C, J = 171, 7, Cp), 84.6 (s, 1 C, C-6), 79.0 [s, 1 C,
C(CH3)3], 65.9 (t, 1 C, J = 140, C-5), 58.8 (s, 1 C, C-2), 58.5 (d,
1 C, J = 132, CHPh), 45.2 (t, 1 C, J = 132, C-1), 42.6 [d, 1 C,
J = 127, CH(CH3)2], 32.1 [q, 3 C, J = 125, C(CH3)3], 24.7 (q,
Ti(C H )(NButCCH CMe᎐CHCMeCH CHPhNPri)(But-
᎐
᎐
3
1 C, J = 128, CH C᎐), 24.2 (q, 1 C, J = 125, CH C᎐), 21.9 (q,
3
᎐
5
5
2
2
1 C, J = 126, CH3), 21.6 (q, 1 C, J = 126, CH3), 21.0 (q, 1 C,
J = 125, CH3), 16.2 [q, 1 C, J = 124, CH(CH3)2] and 15.6 [q, 1 C,
J = 125 Hz, CH(CH3)2]. Mass spectrum (EI, 25 eV): m/z (rela-
tive intensity) 782 (8), 781 (20), 780 (34), 779 (14), 778 (16), 738
(9), 737 (17), 706 (15), 634 (18), 633 (49), 632 (100), 631 (25),
630 (27), 596 (10), 594 (14), 578 (9), 577 (24), 576 (47), 520 (12),
489 (13), 392 (14), 328 (16) and 148 (9). IR (Nujol mull): 3332m
br, 3000–2800vs, 1608w, 1502m, 1460vs, 1377s, 1303w, 1261m,
1096w, 1020w and 806vs.
NC) 5. To a stirred solution of compound 2b (0.50 g, 1.4 mmol)
in 50 mL ether under a nitrogen atmosphere at Ϫ78 ЊC was
added tert-butyl isocyanide (0.32 mL, 2.8 mmol). A rapid
change from dark to bright red occurred upon the addition.
The reaction mixture was then warmed with stirring to 0 ЊC and
the solvent removed in vacuo to leave behind a dark red solid.
The solid was in turn extracted with three 50 mL portions of
ether–pentane (50:50) and the extracts were filtered through a
Celite pad on a coarse frit. The resulting bright red filtrate was
concentrated in vacuo to ca. 20 mL and placed into a Ϫ30 ЊC
freezer for one week. The supernatant was removed via syringe,
and the solid dried in vacuo yielding 0.31 g (42%) of the product
as air-sensitive dark red crystals (mp 117–119Њ) (Found: C,
73.43; H, 9.08; N, 8.05. Calc. for C32H47N3Ti: C, 73.68; H, 9.08;
Ti(C H )(NPriCHPhCH CMe᎐CHCMe᎐CH )Cl 7. Into a
᎐
᎐
5
5
2
2
2
NMR tube was added compound 26 (60 mg, 0.20 mmol) and
C2Cl6 (40 mg, 0.20 mmol) followed by 0.60 mL of deuteriated
benzene. After sealing the tube the mixture was shaken during
which a bright red color was observed. Isolation of the com-
pound using the same general procedure as that for the above
complex gave a red oil. 1H NMR (benzene-d6, ambient): δ 7.51–
6.89 (m, 5 H, Ph), 6.21 (s, 5 H, Cp), 5.63 (s, 1 H, H-3), 5.01 (s,
1 H), 4.78 (s, 1 H), 4.76 (d, 1 H, J = 11, H-5), 3.97 (dd, 1 H,
J = 11, 15, CHPh), 3.30 [septet, 1 H, J = 6.3, CH(CH3)2], 2.26
1
N, 8.05%). H NMR (benzene-d6, ambient): δ 7.42–7.01 (m,
5 H, Ph), 6.09 (s, 5 H, Cp), 5.77 (s, 1 H, H-3), 5.72 (dd, 1 H,
(J = 5, 8), 3.33 (d, 1 H, J = 16, H-5), 3.30 [septet, 1 H, J = 6.6,
CH(CH3)2], 2.52 (dd, 1 H, J = 8, 14), 2.00 (dd, 1 H, J = 5, 14),
1.86 (s, 3 H, CH C᎐), 1.70 (d, 1 H, J = 16, H-5Ј), 1.44 [s, 9 H,
᎐
3
(CH ) CNC], 1.20 (s, 3 H, CH C᎐), 1.10 [d, 3 H, J = 6.8,
᎐
3
3
3
(d, 1 H, J = 15, H-5Ј), 1.75 (s, 3 H, CH ), 1.63 (s, 3 H, CH C᎐),
CH(CH3)2], 0.92 [d, 3 H, J obscure, CH(CH3)2] and 0.91 [s, 9 H,
(CH3)3CNC]. 13C NMR (toluene-d8, Ϫ40 ЊC): δ 151.6 (s, 1 C,
CN), 141.0 (s, 1 C, C-4), 140.0 (s, 1 C, Ph), 109.8 (d, 1 C,
J = 170, C-3), 108.0 (d of quintets, 5 C, J = 170, 6, Cp), 88.4 (s,
1 C, C-6), 67.5 (d, 1 C, J = 135, CHPh), 61.0 (d, 1 C, obscured,
CH(CH3)2], 59.8 (s, 1 C, C(CH3)3], 56.5 (s, 1 C, C-2), 53.0 [s, 1 C,
C(CH3)3], 47.6 (t, 1 C, J = 125, C-5), 43.0 (t, 1 C, J = 126, C-1),
᎐
3
3
1.20 (d, 3 H, J = 6.4, CH3) and 1.15 (d, 3 H, J = 6.1 Hz, CH3).
13C NMR (benzene-d6, ambient): δ 157.6 (s, 1 C), 144.4 (s, 1 C),
142.1 (s, 1 C), 134.9–128.0 (m, 5 C), 118.2 (d of quintets, 5 C,
J = 177, 6, Cp), 114.9 (t, 1 C, J = 156, C-1), 65.3 (d, 1 C, J = 132,
CHPh), 51.3 [d, 1 C, J = 127, CH(CH3)2], 41.3 (t, 1 C, J = 128,
C-5), 23.9 (q, 1 C, J = 126, CH C᎐), 23.2 (q, 1 C, J = 125,
᎐
3
CH C᎐), 20.7 [q, 1 C, J = 127, CH(CH ] and 19.9 [q, 1 C,
34.0 [q, 3 C, J = 129, C(CH ) ], 33.8 (q, 1 C, J = 129, CH C᎐),
᎐
32
J = 3128 Hz, CH(CH3)2].
᎐
3
3
3
29.0 (q, 1 C, J = 131, CH C᎐), 28.8 [q, 3 C, J = 131, C(CH ) ],
᎐
3
3 3
26.6 [q, 1 C, J = 127, CH(CH3)2] and 24.4 [q, 1 C, J = 127 Hz,
CH(CH3)2]. Mass spectrum (EI, 70 eV): m/z (relative intensity)
521 (3), 520 (2), 357 (12), 208 (29), 206 (27), 188 (29), 169 (10),
166 (11), 165 (100), 148 (14), 147 (14), 132 (16), 109 (19), 57
(21), 43 (15) and 41 (15). IR (Nujol mull): 3000–2800vs, 2172vs,
1600vw, 1460vs, 1376vs, 1370w, 1261vs, 1240m, 1095s, 1063m,
1019vs, 910w, 798vs and 701s.
Ti(C5H5)(NPriCHPhCH2CMeCHCMe2)(ButO) 8. To
a
stirred solution of compound 2b (0.50 g, 1.4 mmol) in 50 mL
ether under an atmosphere of nitrogen at Ϫ78 ЊC was added
tert-butyl alcohol (0.13 mL, 1.4 mmol). A slow change from
dark to light red occurred upon warming to room temperature.
The mixture was stirred at room temperature for 1 h. The solv-
ent was next removed in vacuo, leaving a dark red sticky oil
which was was extracted with three 50 mL portions of pentane.
These extracts were filtered through a Celite pad on a coarse
Compound 6. To a stirred solution of compound 2b (0.50 g,
1.4 mmol) in 50 mL THF under nitrogen at Ϫ78 ЊC was added a
J. Chem. Soc., Dalton Trans., 1999, 1883–1890
1889

View Article Online
Table 5 X-Ray data parameters for compounds 3, 4, 5 and 6
3
4
5
6
Formula
M
C28H34N2Ti
446.50
C25H35NOTi
413.46
C32H47N3Ti
521.63
C54H70N4O2Ti
855.09
Space group
Crystal system
T/K
Pbca
Orthorhombic
193
P21/c
Monoclinic
193
P21/c
Monoclinic
193
P21/c
Monoclinic
295
a/Å
b/Å
c/Å
8.620(3)
23.051(9)
23.771(7)
13.650(3)
8.995(2)
19.006(6)
105.33(2)
2250.6
12.626(3)
14.766(4)
64.305(3)
97.71(2)
3012.4
11.286(1)
10.940(4)
41.288(7)
90.00(1)
5098.1
β/Њ
V/Å3
4723.5
Z
8
4
4
4
Crystal dimensions/mm
0.49 × 0.25 × 0.15
3.74
0.25 × 0.25 × 0.09
3.89
0.36 × 0.29 × 0.22
3.08
0.22 × 0.18 × 0.07
2.04
µ/cmϪ1
Data collected
Unique data with I > nσ(I); n
R(F)
4253
2272; 3
0.046
0.054
0.30
3959
2300; 3
0.038
0.047
0.24
5527
8813
2128; 3
0.096
0.098
0.57
5274; 2
0.041
RЈ(F)
0.088a
Maximum difference peak/e ÅϪ3
0.29
a SHELXS definition.
frit. The bright red filtrate was concentrated in vacuo to ca. 20
mL and placed into a Ϫ90 ЊC freezer for 2 weeks during which
a red oil was formed. The supernatant was removed via syringe,
and the oil pumped on for several hours with no apparent
References
1 E. Meléndez, A. M. Arif, R. D. Ernst and M. L. Ziegler, Angew.
Chem., Int. Ed. Engl., 1988, 27, 1099.
2 T. E. Waldman, A. M. Wilson, A. L. Rheingold, E. Meléndez and
R. D. Ernst, Organometallics, 1992, 11, 3201.
1
change. H NMR (benzene-d6, ambient): δ 7.35–7.06 (m, 5 H,
3 A. M. Wilson, T. E. Waldman, A. L. Rheingold and R. D. Ernst,
J. Am. Chem. Soc., 1992, 114, 6252.
4 R. Bloch, Chem. Rev., 1998, 98, 1407; S. L. Buchwald, B. T. Watson,
M. W. Wannamaker and J. C. Dewan, J. Am. Chem. Soc., 1989, 111,
4486.
5 J. C. Huffman, K. G. Moloy, J. A. Marsella and K. G. Caulton,
J. Am. Chem. Soc., 1980, 102, 3009.
6 A. F. Wells, Structural Inorganic Chemistry, Oxford University Press,
Oxford, 1984.
Ph), 6.13 (s, 5 H, Cp), 5.60 (s, 1 H, H-3), 4.31 (dd, 1 H, J = 7, 12,
H-5), 2.66 [septet, 1 H, J = 6.6, CH(CH3)2], 2.60 (dd, 2 H,
J = 12, 15, H-5Ј), 2.09 (dd, 1 H, J = 7, 15, CHPh), 2.00 (s, 3 H,
CH3), 1.78 (s, 3 H, CH3), 1.65 (s, 3 H, CH3), 1.39 [s, 9 H,
(CH3)3C], 1.25 [d, 3 H, J = 6.1, (CH3)2CH] and 0.82 [d, 3 H,
J = 6.9 Hz, CH3)2CH]. 13C NMR (benzene-d6, ambient): δ 147.1
(s, 1 C, Ph), 144.5 (d, 1 C, J = 142, C-3), 128.7–126.7 (m, 5 C,
Ph), 121.4 (s, 1 C, C-4), 113.0 (d of quintets, 5 C, J = 171, 7,
Cp), 83.6 (s, 1 C, C-2), 83.0 [s, 1 C, C(CH3)3], 61.5 (d, 1 C,
J = 135, CHPh), 53.9 [d, 1 C, J = 132, CH(CH3)2], 49.4
(t, 1 C, J = 128, C-1), 32.8 [q, 3 C, J = 126, C(CH3)3], 27.2 (q,
1 C, J = 127, CH C᎐), 26.6 (q, 1 C, J = 127, CH C᎐), 25.0
7 V. Schomaker and D. P. Stevenson, J. Am. Chem. Soc., 1941, 63, 37.
8 D. Jacoby, S. Isoz, C. Floriani, A. Chiesi-Villa and C. Rizzoli, J. Am.
Chem. Soc., 1995, 117, 2805.
9 See G. A. Stark and J. A. Gladysz, Inorg. Chem., 1996, 35, 5509.
10 F. A. Cotton, S. A. Duraj and W. J. Roth, J. Am. Chem. Soc., 1984,
106, 6987; C. Valero, M. Grehl, D. Wingbermühle, L. Kloppenburg,
D. Carpenetti, G. Erker and J. Petersen, Organometallics, 1994, 13,
415; P. L. Motz, J. J. Alexander and D. M. Ho, Organometallics,
1989, 8, 2589; J. R. Bocarsly, C. Floriani, A. Chiesi-Villa and
C. Guastini, Organometallics, 1986, 5, 2380; B. Hessen, J. Blenkers,
J. H. Teuben, G. Helgesson and S. Jagner, Organometallics, 1989, 8,
830; F. J. Berg and J. L. Petersen, Organometallics, 1989, 8, 2461;
C. P. Gerlach and J. A. Arnold, Organometallics, 1996, 15, 5260.
11 R. Tomaszewski and R. D. Ernst, unpublished results.
12 B. D. Steffey, P. E. Fanwick and I. P. Rothwell, Polyhedron, 1990, 9,
963; P. T. Wolczanski, Polyhedron, 1995, 14, 3335; W. A. Howard,
T. M. Trnka and G. Parkin, Inorg. Chem., 1995, 34, 5900.
13 W. W. Lukens, Jr., M. R. Smith III and R. A. Andersen, J. Am.
Chem. Soc., 1996, 118, 1719.
14 R. M. Laine, R. E. Moriarty and R. Bau, J. Am. Chem. Soc., 1972,
94, 1402; R. B. King, Inorg. Chem., 1968, 7, 1044.
15 H. Yasuda and A. Nakamura, Angew. Chem., Int. Ed. Engl., 1987,
26, 723; see also J. A. Hanna, A. M. Baranger, P. J. Walsh and
R. G. Bergman, J. Am. Chem. Soc., 1995, 117, 3292.
16 I. Ugi and E. Meyr, Org. Synth., 1973, Coll. Vol. V, 1060.
17 I. Hyla-Kryspin, T. E. Waldman, E. Meléndez, W. Trakarnpruk,
A. M. Arif, M. L. Ziegler, R. D. Ernst and R. Gleiter,
Organometallics, 1995, 14, 5030.
᎐
᎐
3
3
(q, 1 C, J = 126), 20.8 [q, 1 C, J = 122, CH(CH3)2] and 19.7 [q,
1 C, J = 125 Hz, CH(CH3)2].
X-Ray crystallography
Single crystals of compounds 3, 4, 5, and 6 were grown by
slowly cooling their concentrated solutions in hydrocarbon
solvents (ether for 6) to Ϫ30 ЊC. These were then mounted in
glass capillaries under a nitrogen atmosphere, and transferred
to a Nonius CAD-4 diffractometer for data collection. Pertin-
ent data collection and refinement parameters are contained
in Table 5. Initial structural solutions were obtained using
direct methods, which located the titanium and most other
non-hydrogen atoms. The remaining non-hydrogen atoms,
including for 6 those of a molecule of ether which was present
in the lattice, were thereafter obtained from Fourier-difference
maps. Hydrogen atoms were then included in idealized posi-
tions, but their atomic coordinates not refined. The SHELXS
program was utilized for 5, MOLEN for the other structures.19
CCDC reference number 186/1411.
18 T. D. Newbound, L. Stahl, M. L. Ziegler and R. D. Ernst,
Organometallics, 1990, 9, 2962.
19 G. M. Sheldrick, SHELXS 86, University of Göttingen, 1986;
C. K. Fair, MOLEN, Enraf-Nonius, Delft, 1990.
Acknowledgements
R. D. E. is grateful to the University of Utah and the National
Science Foundation for partial support of this work. We thank
Dr. Charles L. Mayne for helpful NMR discussions.
Paper 9/01304K
1890
J. Chem. Soc., Dalton Trans., 1999, 1883–1890