Alkane binding implicated in reactions of (tBu 3SiN=)3WHK and alkyl halides

Article abstract of DOI:10.1021/om2009342

The reduction of alkyl, benzyl, allyl, and propargyl halides by [( ^tBu₃SiN=)₃WH]K (2-H-M) was assayed via analysis of the product distributions. For unhindered sp³ substrates (MeX, EtX, BnX, CH₂=CHCH₂X, RCCCH₂X, etc.), the observed products were RH loss and [(^tBu₃SiN=) ₃WX]M (2-X-M) or MX elimination and (^tBu ₃SiNH)(^tBu₃SiN=)₂WR (1-R), which was derived from 1,2-RH-addition to an imide of the intermediate alkane (arene) complex (^tBu₃SiN=)₃W(RH) (2-RH). Solvent activation or binding was also seen, consistent with the trapping of ( ^tBu₃SiN=)₃W (2) generated from RH loss from 2-RH. The radical clocks ^cPrCH₂Br and 5-hexenyl bromide yielded the unrearranged products (^tBu₃SiNH)( ^tBu₃SiN=)₂W(trans-^c(CHCH ₂CHMe)) (1-^cPrMe) and (^tBu₃SiNH) (^tBu₃SiN=)₂W(trans-CH=CHⁿBu) (1-CH=CHⁿBu), respectively. When hindered sp³ substrates were employed, halogenated products [(^tBu₃SiN=) ₃WX]M (2-X-M) and (^tBu₃SiNH)( ^tBu₃SiN=)₂WX (1-X) were prominent, suggesting that radical paths were operable. KIEs derived from product ratios involving CD₃I or CD₃CH₂I and 2-H-K supported the intermediacies of (^tBu₃SiN=)₃W(CHD₃) (2-CHD₃) and (^tBu₃SiN=)₃W(CH ₃CD₃) (2-CH₃CD₃), respectively. X-ray crystal structures of (^tBu₃SiNH)(^tBu ₃SiN=)₂WR (1-R; R = Me, ^cPrMe, CH=C=CH ₂) are presented, and the results are placed in the context of 1,2-RH-addition to d⁰ early-transition-metal imido species.

Full text of DOI:10.1021/om2009342

Article
pubs.acs.org/Organometallics
Alkane Binding Implicated in Reactions of (tBu3SiN)3WHK and Alkyl
Halides
Daniel F. Schafer, II, Peter T. Wolczanski,* and Emil B. Lobkovsky
Department of Chemistry & Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York 14853, United States
S
* Supporting Information
ABSTRACT: The reduction of alkyl, benzyl, allyl, and
propargyl halides by [(tBu3SiN)3WH]K (2-H-M) was
assayed via analysis of the product distributions. For
unhindered sp3 substrates (MeX, EtX, BnX, CH2
CHCH2X, RCCCH2X, etc.), the observed products were RH
loss and [(tBu3SiN)3WX]M (2-X-M) or MX elimination and (tBu3SiNH)(tBu3SiN)2WR (1-R), which was derived from 1,2-
RH-addition to an imide of the intermediate alkane (arene) complex (tBu3SiN)3W(RH) (2-RH). Solvent activation or binding
was also seen, consistent with the trapping of (tBu3SiN)3W (2) generated from RH loss from 2-RH. The radical clocks
cPrCH2Br and 5-hexenyl bromide yielded the unrearranged products (tBu3SiNH)(tBu3SiN)2W(trans-c(CHCH2CHMe))
(1-cPrMe) and (tBu3SiNH)(tBu3SiN)2W(trans-CHCHnBu) (1-CHCHnBu), respectively. When hindered sp3 substrates
were employed, halogenated products [(tBu3SiN)3WX]M (2-X-M) and (tBu3SiNH)(tBu3SiN)2WX (1-X) were prominent,
suggesting that radical paths were operable. KIEs derived from product ratios involving CD3I or CD3CH2I and 2-H-K supported
the intermediacies of (tBu3SiN)3W(CHD3) (2-CHD3) and (tBu3SiN)3W(CH3CD3) (2-CH3CD3), respectively. X-ray
c
crystal structures of (tBu3SiNH)(tBu3SiN)2WR (1-R; R = Me, PrMe, CHCCH2) are presented, and the results are
placed in the context of 1,2-RH-addition to d0 early-transition-metal imido species.
( t B u 3 S i N H ) 3 T i R , 2 2 ( t B u 3 S i N H ) 3 Z r R , 2 3 , 2 4
INTRODUCTION
■
( t B u 3 S i N H ) 2 ( t B u 3 S i N  ) V R , 2 5 , 2 6 a n d
(tBu3SiNH)2(tBu3SiN)TaR27 systems. Moreover, for related
tungsten imido complexes, 1,2-RH-elimination from
(tBu3SiNH)(tBu3SiN)2WR (1-R) was not observed (Scheme
1). As previously shown, extended thermolysis of (tBu3SiNH)-
(tBu3SiN)2WCD3 (1-CD3) in C6H6 at 200 °C revealed no
H/D exchange between amide and methyl positions, nor was
any solvent activation indicated, consistent with calculations on
Evidence for the intermediacy of alkane complexes in reactions
that ultimately cleave carbon−hydrogen bonds is compelling in
several d6 and d8 systems. The direct observation of a methane
complex of rhodium, [{2,6-(tBu2PO)C5H3N}Rh(CH4)]-
BArF ],1 is a recent highlight that followed related low-
4
temperature NMR spectroscopic characterizations of cyclo-
pentane2 and cyclohexane3 adducts of rhenium, i.e., CpRe-
(CO)2(cAlkH), among others.4−6 Previously, indirect character-
izations of M(RH) species by dynamic NMR spectroscopy
were implicated in the interconversion of hydride and methyl
resonances in tungsten,7 rhenium,8 osmium,9 and iridium10
complexes and via chain-walking in Cp*Rh(PMe3)R(H)
species,11 where the alkane adduct was a logical intermediate
in oxidative addition/reductive elimination events. While not
precursors to oxidative addition, a heptane bound within a
porphyrin pocket12 and several cyclic alkanes embedded within
the arms of a tacn ligand, i.e., [κ-O3,N3-(2,4-(tBu)C6H2O(6-
CH2)tacn]U(RH),13 have been crystallographically character-
ized.
(NH2)(HN)2WCH3 that suggested ΔH⧧ was ∼60 kcal/
elim
mol.28−30
While the activation energy pertaining to 1,2-elimination of
MeH from (tBu3SiNH)(tBu3SiN)2WCH3 (1-CH3)or the
corresponding calculated elimination from (NH2)-
(HN)2WCH3appeared prohibitive with regard to the
observation of a methane adduct, its binding energy was
expected to be substantial on the basis of a calculated ΔG°bind
value of −8.4 kcal/mol (ΔH°bind ≈ −15.6 kcal/mol) for
(HN)3W(CH4).17,18 A significant experimental break
occurred when anionic hydride adducts of “(tBu3SiN)3W
(2)” were synthesized via the methodology previously
delineated.28−30 While [(tBu3SiN)3WH]K (2-H-K) pos-
sesses the solubility properties of a monomer, Figure 1
illustrates its solid-state structure, which is a C2h-symmetric
dimer containing potassiums bridging anionic
[(tBu3SiN)3WH]− units.29,30 As Scheme 1 illustrates, the
treatment of 2-H-K with alkyl halides provided hints of alkane
In d0 carbon−hydrogen bond activation systems where 1,2-
RH-addition to a metal−imido linkage of XYMNR
constitutes the activation event,14 the intermediacy of alkane
complexes was implicated by calculations15−18 but unobserved
by spectroscopy. Observations of 1,2-RH elimination from
XY(R′NH)M−R and corresponding calculations were in
concert in suggesting that dynamic exchange of the amide
hydrogen with hydrogens on the alkyl was energetically
unfeasible. Competitive loss of RH was too swift with respect
to alkane readdition in (tBu3SiO)2(tBu3SiNH)TiR,19−21
Received: October 5, 2011
Published: November 29, 2011
© 2011 American Chemical Society
6518
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 1
[(tBu3SiN)3WH]K (2-H-K) with MeI in THF-d8 generated
∼12% of (tBu3SiNH)(tBu3SiN)2WCH3 (1-CH3), a result
made consistent with a methane complex intermediate. A
switch to benzene as solvent rendered this possibility more
tenable, as heating 2-H-K with 10 equiv of CH3I in C6D6 at 60
°C for 6 h produced 1-CH3 in 90% yield by 1H NMR
spectroscopy, accompanied by 10% (tBu3SiND)-
(tBu3SiN)2WC6D5 (1-ND-C6D5), representing activation of
the solvent, trace amounts of [(tBu3SiN)3WI]K (2-I-K, <2%)
and CH4, and a precipitate, presumed to be KI (Scheme 2).
Use of CD3I as the substrate (10 equiv) afforded ∼70% 1-CD3
and (tBu3SiND)(tBu3SiN)2WCD2H (1-ND-CD2H), an
increased amount of 1-ND-C6D5 (22%), and a measurable
amount (∼8%) of [(tBu3SiN)3WI]K (2-I-K) along with trace
CD3H. It proved difficult to obtain the ratio 1-CD3/1-ND-
CD2H by 1H NMR spectroscopy due to problems assessing the
magnitude of the −CD2H quintet at the foot of a much larger
resonance. A decline in the integrated NH resonance relative to
the 54 H imide methyl resonance from 0.86(2)/54 vs 1.02(2)/
54 in samples of 1-CH3 was consistent with the formation of
some 1-ND-CD2H. To check for the possibility of exogenous
methane activation, the experiment was repeated with 5 equiv
of CH4 present, and only CD3H-derived products were
generated (∼65%) along with 9% of 2-I-K. Little if any 1-
CH3 was detected (<5%), consistent with methane activation
solely from CD3H. The potassium iodide adduct 2-I-K is not
the origin of methane or benzene solvent activation, as
independent thermolyses confirmed its stability. What is not
clear is whether solvent activation occurs during the alkylation
process or whether it is due to the capture of putative
(tBu3SiN)3W (2) formed via MeH escape from 2-MeH.
The ratio 1-CD3/1-ND-CD2H was determined from a
separate experiment conducted in C6H6, so that 2H{1H}
NMR could be used for the assay. Under the same conditions,
73% 1-CD3 and 1-ND-CD2H were obtained, with 27%
(tBu3SiNH)(tBu3SiN)2WPh (1-Ph) and <2% 2-I-K. The
relative integrated intensity of the methyl resonance at δ 1.41 to
the amide at δ 6.60 was 11.5(8). Assuming the only two
contributing molecules to these resonances are 1-CD3 and 1-
ND-CD2H, the ratio 1-CD3/1-ND-CD2H was 3.2(2)/1, and
after multiplying by the statistical factor of 3, the kH/kD value
for intramolecular methane activation via putative
Figure 1. Ball and stick view of the solid-state dimer of 2-H-K: i.e.,
[(tBu3SiN)3WH]2(μ-K)2 ((2-H)2(μ-K)2).
complex intermediates, as the transfer of hydride from 2-H-K to
RX provided a separate, indirect path to these transient species.
In essence, the results implied that the free energy well in which
(tBu3SiN)3W(RH) (2-RH) existed was deep enough that
CH activation to give (tBu3SiNH)(tBu3SiN)2WR (1-R)
could be competitive with alkane loss in certain instances.
Detailed investigations into the probable intermediacy of alkane
complexes are delineated below; a portion of this effort has
been communicated.28
RESULTS
■
[(tBu3SiN)3WH]K (2-H-K) and RX. As a hydride source,
[(tBu3SiN)3WH]K (2-H-K) proved most convenient to
synthesize, and it was used almost exclusively in these
investigations. Reactions were typically run with RX in excess,
i.e., pseudo-first-order conditions, for reasonable conversions.
Many of the products, such as (tBu3SiNH)(tBu3SiN)2WR (1-
R ) , ( t B u 3 S i N H ) ( t B u 3 S i N  ) 2 W X ( 1 - X ) , a n d
[(tBu3SiN)3WX]K (2-X-K), have been previously synthe-
sized and characterized;30 NMR spectroscopic data for any new
compounds are given in Table 1.
[(tBu3SiN)3WH]M (2-H-M) and MeX. 1. Product
Distributions. As previously reported,30 treatment of
6519
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
a
1
Table 1. H and 13C{1H} NMR Spectral Assignments for (tBu3NH)(tBu3SiN)2WX/R (1-X/R)and (tBu3SiN)3WL
1H NMR (δ (mult, J (Hz), assignt))
13C{1H} NMR (δ (J (Hz), assignt))
b
tBu
b
b
compd
NH and R
0.70 (t, 7, CH3)
SiCMe3 SiC(CH3)3
R
(tBu3SiN)3WOEt2 (2-OEt2)
1.40
1.42
24.40
31.55
12.80 (CH3)
80.29 (CH2)
26.01 (CH2)
85.91 (OCH2)
21.92 (C4H3)
21.92 (C2H)
4.09 (q, 7, CH2)
0.96 (m, CH2)
4.16 (m, OCH2)
6.30 (NH)
1.06 (3.53c3t, 5.123t, 9.213t
C3Ht)
(tBu3SiN)3W(THF) (2-THF)
24.33
31.54
cd
,
(tBu3SiN)2(tBu3SiNH)WcPrMe (1-cPrMe)
1.24
1.35
23.44
24.40
30.79
31.22
,
22.97 (C3H)
49.48 (1JWC = 187,
1.14 (5.924, C4H3)
C1H)
1.65 (5.312, 6.513c, 9.213t
C1H)
,
1.89 (3.53c3t, 6.513c, 7.123c
,
C3Hc)
2.09 (5.123t, 5.312, 5.924
7.123c, C2H)
(tBu3SiN)2(tBu3SiNH)W(CHCEt) (1-CH
e
6.76 (NH)
CHEt)
6.94 (dt, 17, 6.5, CHEt)
7.90 (d, 17, WCH)
6.62 (NH)
(tBu3SiN)2(tBu3SiNH)W(CHCnBu) (1-CH
1.24
1.36
23.57
24.54
30.80
31.26
22.37 (CH3)
f
CHnBu)
0.84 (t, 7, CH3)
24.95 (CH2)
31.89 (CH2)
38.50 (CH2)
158.90 (CH)
166.19 (CH)
g
2.06 (m, CHCH2)
8.01 (d, 17, WCH)
(tBu3SiNH)(tBu3SiN)2W(CH2C6H4-p-Me) (1-xyl)
1.22
1.29
6.71 (NH)
23.38
30.81
2.20 (CH3)
3.59 (JWH = 15, CH2)
7.01 (d, 7, CH)
7.24 (d, 7, CH)
6.83 (NH)
(tBu3SiNH)(tBu3SiN)2W(C6H3-2,5-Me2) (1-2,5-
1.21
1.37
23.53
24.70
30.93
31.51
20.21 (CH3)
h
Me2Ph)
2.21 (CH3)
24.49 (CH3)
129.42 (Ar)
130.75 (Ar)
133.17 (Ar)
142.62 (Ar)
2.85 (CH3)
6.90 (d, 8, CH)
8.71 (s, CH)
148.08 (Cipso
21.63 (CH3)
)
)
)
(tBu3SiNH)(tBu3SiN)2W(2-C10H6-6-Me)(1-6-Me-2-
1.21
1.40
6.95 (NH)
23.57
24.59
30.80
31.33
i
nap)
2.16 (CH3)
125.73 (Ar)
127.37 (Ar)
128.49 (Ar)
132.29 (Ar)
134.15 (Ar)
136.46 (Ar)
139.18 (Ar)
144.95 (Ar)
171.97 (Cipso
21.63 (CH3)
7.09 (d, 8, CH)
7.28 (CH5)
7.61 (d, 8, CH)
7.78 (d, 8, CH)
8.50 (d, 8, CH)
9.15 (CH1)
(tBu3SiNH)(tBu3SiN)2W(2-C10H6-7-Me) (1-7-Me-2- 1.21
6.97 (NH)
23.57
24.59
30.80
30.80
i
nap)
1.40
2.18 (CH3)
126.09 (Ar)
127.15 (Ar)
129.10 (Ar)
132.20 (Ar)
134.26 (Ar)
135.78 (Ar)
138.35 (Ar)
144.37 (Ar)
172.90 (Cipso
7.02 (d, 8, CH)
7.46 (d, 8, CH)
7.61 (d, 8, CH)
7.69 (CH8)
8.46 (d, 8, CH)
9.13 (CH1)
6520
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Table 1. continued
1H NMR (δ (mult, J (Hz), assignt))
13C{1H} NMR (δ (J (Hz), assignt))
b
tBu
b
b
compd
NH and R
6.71 (NH)
SiCMe3 SiC(CH3)3
R
(tBu3SiNH)(tBu3SiN)2W(CH2C6H3-3,5-Me2) (1-
1.23
1.29
23.39
30.83
21.38 (CH3)
mes)
2.19 (CH3)
24.49
31.14
50.92 (CH2)
126.62 (Ar)
127.66 (Ar)
137.54 (Ar)
146.11 (Ar)
3.58 (JWH = 15, CH2)
6.55 (CH4)
6.98 (CH2,6
)
(tBu3SiNH)(tBu3SiN)2W(CHCCH2) (1-CH
CCH2)
1.22
1.33
6.87 (NH)
23.53
24.54
30.76
31.24
67.08 (3JWC = 7,
CH2)
4
4.35 (d, 7, JWH = 7, CH2)
2
6.70 (t, 7, JWH = 7, CH)
109.08 (1JWC = 179,
CH)
212.07 (C)
(tBu3SiNH)(tBu3SiN)2WSiMe3 (1-SiMe3)
1.19
1.34
1.17
6.97 (NH)
23.10
24.13
23.52
30.64
31.24
30.89
10.08 (CH3)
0.81 (CH3)
2.62 (d, 2, 3JWH = 13, CH3)
j
j
j
j
l
(tBu3SiN) W(κ-N,C-N(SitBu3)CHC(Me)−) (3-
k2
HCCMe)
3
1.32
1.22
5.89 (q, 2, JWH = 54, CH)
24.23
31.06
31.26
l
(tBu3SiN)2W(κ-N,C-N(SitBu3)CMeCMe−) (3-
2.28 (CH3)
24.8
15.81 (3JWC = 26,
CH3)
MeCCMe)
1.32
2.52 (3JWH = 13, CH3)
24.32
31.13
20.71 (4JWC = 9,
CH3)
126.23 (2JWC = 15,
WCC)
174.24 (1JWC = 161,
WC)
(tBu3SiN)2W(κ-C,η3-C,N-CH2CMe2Si(tBu2)
NCMeCHMe) (4-MeCCMe)
1.27
1.37
1.04 (SitBu)
1.21 (SitBu)
21.30
24.64
31.20
31.36
16.66 (C(CH3)Me)
22.91 (CMe(CH3))
23.41 (SiC)
1.48 (C(CH3)Me)
1.50 (CMe(CH3)
1.82 (d, 5.5, NCCCH3
2.31 (NCCH3)
24.46 (SiC)
33.54 (NCCH3)
33.79 (NCCCH3)
35.29 (CH2CMe2)
2
2.60 (d, 13, JWH = 10,
WCHH)
2
2.75 (d, 13, JWH = 10,
WCHH)
53.28 (JWC = 24,
NCC)
3.18 (q, 5.5, NCCH)
54.08 (1JWC = 105,
WCH2)
180.47 (2JWC = 10,
NC)
a
In benzene-d6 unless otherwise noted. All JWH values reported were obtained from W satellites that integrated to ∼14% intensity, indicative of one
b
c
tungsten center. The amide is listed first and the imide second. From E.COSY and HMQC experiments: W-c(C1HC2H(C4H3)C3HcHt), e.g., 3J13t
=
d
e
9.2 Hz listed as 9.213t
.
1JCH = 145 (C1H), 163 (C2H), 161 (C3Hc), 160 (C3Ht), 126 Hz (C4H3). Resonances obscured by the major product
g
f
1-cPrMe. Methylene obscured by the tBu3 group and CH obscured by the residual solvent in 1H NMR. Minor product (14%); hence, aryl carbons
h
i
are not observed. Aryl resonance obscured by residual solvent; one CMe not observed. One aryl carbon not observed; assignments tentative due to
j
k
overlapping resonances, and similar intensities with isomer. Metallacyclic NSitBu3 listed first. Since the metallacycle is a minor product, carbon
l
resonances are not identified with confidence. Metallacyclic NSitBu3 resonances listed first; both are broad.
(tBu3SiN)3W(CD3H) (2-CD3H) was 9.6(6). As previously
stated, the isotopomers do not thermally interconvert; hence,
the ratio reflects a kinetic product distribution. Herein it is also
assumed that differential binding of the methane, e.g.,
(tBu3SiN)3W(η2-HDCD2) vs (tBu3SiN)3W(η2- D2CHD),
is fast and reversible; hence, the KIE reflects the relative
transition states for CH vs CD activation. Nonetheless, the KIE
must be considered a phenomenological number reflecting
both binding and activation events.
H-K(c-2.2.2)),30 where the K+ is complexed by crypt-2.2.2, was
used in a CH3I reaction according to eq 1. The results are quite
similar to the previous trials, with the 76% conversion to
(tBu3SiNH)(tBu3SiN)2WCH3 (1-CH3) comparable to trial
2.1 in Scheme 2. No evidence of uncomplexed crypt-2.2.2 was
1
obtained by H NMR spectroscopic analysis, and the observed
crypt-2.2.2 was attributable to 15% [(tBu3SiN)3WI]K(c-
2.2.2) (2-I-K(c-2.2.2)).
A switch to donor solvents (Scheme 3) for the addition of
CD3I to [(tBu3SiN)3WH]K (2-H-K) did not change the
amount of methane activation under comparable conditions
(60 °C, 10 h), as 73% (tBu3SiNH)(tBu3SiN)2WCD3 (1-
The potassium counterion may or may not be bound to the
[(tBu3SiN)3WH] core in 2-H-K during the course of methyl
halide addition; therefore, [(tBu3SiN)3WH][K-c-2.2.2] (2-
6521
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 2
(∼68% total) ratio remains essentially the same for both X = Br
and X = I. Use of 2-H-Na (trial 4.2) causes a significant increase
in 2-I-Na vs 2-I-K production in trial 4.1, although the 1-CH3/
1-ND-C6D5 ratio is similar.
Nucleophilic and electron transfer mechanisms for CH3X
reduction should depend on the nature of X, but if the
transition state for this component of the overall process is very
early, it is possible that the product distribution would not be
very sensitive to these changes. Note that the methane vs
solvent activation ratio is relatively constant for all trials,
perhaps indicating that the latter is derived from dissociation of
CH4 from putative transient (tBu3SiN)3W(CH4) (2-CH4). It
is also plausible that the Na counterion has a greater affinity for
heteroatoms in the core of the tungsten hydride; hence,
collapse to the NaI adduct in trial 4.2 is more favorable than in
the related reaction with potassium (trial 4.1). In the reaction of
CH3I with 2-H-Na, a small amount (4%) of (tBu3SiNH)-
(tBu3SiN)2WI (1-I) is also produced, but its origin is not
clear.
(1)
CD3) and (tBu3SiND)(tBu3SiN)2WCD2H (1-ND-CD2H)
formed in diethyl ether. Instead of solvent activation, binding of
Et2O was noted in the form of (tBu3SiN)3W(OEt2) (2-OEt2,
27%), and no 2-K-I was observed. At a higher temperature (100
°C, 75 min), a similar reaction in THF afforded only 33%
methane activation, and the major product was the THF adduct
(tBu3SiN)3W(THF) (2-THF, 45%), accompanied by 22% 2-
I-K. The deuterium incorporation in the amide position was
assayed by 2H{1H} NMR spectroscopy to provide kH/kD values
for HCD3 activation of 9.9(6) in Et2O and 11.7(6) in THF.
Some modest hydride and substrate variations were tested in
C6D6, and the results are given in Scheme 4. One assumption is
with regard to the structure of [(tBu3SiN)3WH]Na (2-H-
Na), which is presumed to be similar to that of 2-H-K. At 60
°C, it appears that the nature of CH3X (X = Cl, Br, I) has little
effect on the product distribution, with 88−90% (tBu3SiNH)-
(tBu3SiN)2WCH3 (1-CH3), 8−10% (tBu3SiND)(tBu3Si−
N)2WC6D5 (1-ND-C6D5), and very little 2-X-K (≤3%)
produced. At lower temperature (23 °C), a greater amount
(∼32%) of 2-X-K is generated while the 1-CH3/1-ND-C6D5
2. Kinetics of [(tBu3SiN)3WH]M (2-H-M) and MeX. Some
limited kinetics data were obtained for reactions of
[(tBu3SiN)3WH]M (2-H-M) and MeX (X = Cl, Br, I) at
60.0(4) °C in C6D6 (Table 2). The kinetics runs were subject
to two experimental problems: (1) the salt derived from the
reduction caused shimming difficulties during 1H NMR
spectroscopic monitoring, and (2) the only integrable signal
was a tungsten hydride satellite resonance of 2-H-M, which
became very weak after ∼2 half-lives. The central W−H
resonance for both 2-H-M (M = Na, K) was obscured by
t
residual solvent, and the signal pertaining to the Bu3Si group
was clustered with the products.
The experiments were conducted under pseudo-first-order
conditions, and plots of ln([2-H-K]/[2-H-K]0) were linear,
whereas second-order plots were decidedly nonlinear. Varia-
6522
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 3
Scheme 4
Table 2. Rate Constants for the Reaction between [(tBu3SiN)3WH]M (2-H-M; M = Na, K) and MeX (X = Cl, Br, I) at 60.0(4)
°C in Benzene-d6
b
a
expt
[2-MH]0 (mM)
[CH3X]0 (mM)
amt of CH3X (equiv)
no. of half-lives
kobs (105 s−1
)
k (104 M−1 s−1
)
(1) 2-H-K + CH3I
(2) 2-H-K + CH3I
(3) 2-H-Na + CH3I
(4) 2-H-Na + CH3I
(5) 2-H-K + CH3Br
(6) 2-H-K + CH3Cl
15
15
15
16
15
11
150
300
150
368
480
10
20
10
23
32
10
3
16.1(6)
25.1(9)
3.6(3)
6.9(2)
43(1)
10.7(4)
8.4(3)
2.4(2)
1.88(5)
9.0(2)
3.5
1
3
3
c
110
1
0.8(1)
0.7(1)
a
b
c
Assumes the order in CH3X is 1: i.e., kobs = k[CH3X]. Average of three NMR tube experiments run simultaneously, unless noted otherwise. Single
NMR tube experiment.
tions of [CH3I] in runs 1−4, while limited, provided data
consistent with a non-zero-order dependence on [CH3I];
therefore an order of 1 was assumed for all CH3X. The average
second-order rate constant for CH3I and 2-H-K is [9.7(20)] ×
10−4 M−1 s−1, which was roughly 4−5 times faster than the
average of methyl iodide and the sodium congener 2-H-Na
([2.14(34)] × 10−4 M−1 s−1). Recall that reactions with 2-H-Na
tend to produce more halide adduct (i.e., 2-X-Na), and it may
be that capture of the tungsten center by NaX during the
reduction process is more competitive than the related
6523
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
scavenging by KX. The single set of runs with 2-H-K and
CH3Br afford a rate constant comparable with that of methyl
iodide, but the limited data on CH3Cl suggest that it is roughly
an order of magnitude slower. This very modest trend I ≥ Br >
Cl is expected for various processes such as nucleophilic
hydride delivery, electron transfer, and even a radical process
and does not provide detailed mechanistic information.31
Structure of (tBu3SiNH)(tBu3SiN)2WCH3 (1-CH3). A
selection of crystallographic and refinement details pertaining
to (tBu3SiNH)(tBu3SiN)2WCH3 (1-CH3) is given in Table
Table 4. Selected Interatomic Bond Distances (Å) and
Angles (deg) for (tBu3SiN)2(tBu3SiNH)WCH3 (1-CH3),
(tBu3SiN)2(tBu3SiNH)W(cCHCH2CHMe) (1-cPrMe), and
(tBu3SiN)2(tBu3SiNH)WCHCCH2 (1-CHCCH2)
a
ab
,
c
1-Me
1-cPrMe
1-CHCCH2
W1−N1
W1−N2
W1−N3
W1−C
1.815(5)
1.823(5)
1.836(5)
2.158(6)
1.765(3)
114.3(2)
113.9(2)
115.8(2)
102.8(2)
103.7(2)
104.2(2)
160.1(3)
161.3(3)
159.9(3)
1.811(14), 1.826(14)
1.825(13), 1.83(2)
1.837(15), 1.82(2)
2.21(4), 2.10(4)
1.782(4)
1.860(4)
1.850(4)
2.155(6)
1.761(6)
111.1(2)
116.5(2)
115.2(2)
101.9(2)
105.6(2)
104.8(2)
162.6(3)
157.3(2)
N−Siav
1.73(3)
N1−W1−N2
N1−W1−N3
N2−W1−N3
N1−W1−C
N2−W1−C
N3−W1−C
W1−N1−Si1
W1−N2−Si2
W1−N3−Si3
113.7(8), 113.4(9)
113.3(8), 112.6(9)
115.6(8), 110.3(11)
103.6(14), 104.2(12)
99.5(13), 105.7(13)
109.2(14), 110.2(14)
161.9(13), 169.4(13)
162.3(12), 160.0(16)
162.7(12), 173.3(15)
Table 3. Selected Crystallographic and Refinement Data for
(tBu3SiN)2(tBu3SiNH)WCH3 (1-CH3),
(tBu3SiN)2(tBu3SiNH)W(cCHCH2CHMe) (1-cPrMe), and
(tBu3SiN)2(tBu3SiNH)WCHCCH2 (1-CHCCH2)
a
1-Me
1-cPrMe
1-CHCCH2
formula
formula wt
space group
Z
C37H85N3Si3W
840.19
C40H89N3Si3W
904.28
C39H85N3Si3W
864.23
157.4(3)
P21/c
P1
P21/n
̅
a
b
4
4
4
Amide and imide ligands are statistically disordered. Two
independent molecules in the asymmetric unit; related parameters
from the two molecules are listed together. Cyclopropyl distances are
d(C1−C2) = 1.46(5) Å, d(C1−C3) = 1.42(5) Å, d(C2−C3) =
1.49(5) Å, and d(C3−C4(Me)) = 1.61(7) Å; angles are C3−C1−C2 =
62.1(29)°, C1−C2−C3 = 57.7(26)°, C1−C3−C2 = 60.3(29)°, C1−
C3−C4(Me) = 100.6(43)°, and C2−C3−C4(Me) = 105.1(41)°.
a, Å
22.077(4)
12.822(3)
17.322(4)
90
12.906(3)
17.735(4)
22.474(5)
76.43(3)
88.86(3)
79.59(3)
4916.8(17)
1.222
12.9530(2)
21.0289(2)
17.0079(2)
90
b, Å
c, Å
α, deg
β, deg
109.91(3)
90
92.2291(9)
90
γ, deg
V, Å3
ρcalcd, g cm−3
μ, mm−1
temp, K
λ (Å)
c
Allenyl distances are d(C1−C2) = 1.183(8) Å and d(C2−C3) =
4610.3(16)
1.209
4629.23(10)
1.241
1.320(11) Å; the W−C1−C2 angle is 125.0(5)°, and the C1−C2−C3
angle is 178.0(10)°.
2.609
2.451
2.600
293(2)
0.71073
R1 = 0.0346
293(2)
173(2)
0.71073
R1 = 0.1083
0.71073
R1 = 0.0423
R indices (I >
2σ(I))
bc
,
wR2 = 0.0730
R1 = 0.0578
wR2 = 0.2398
R1 = 0.2432
wR2 = 0.0686
R1 = 0.1044
R indices (all
bc
,
data)
wR2 = 0.0843
1.040
wR2 = 0.3537
1.045
wR2 = 0.1005
1.103
d
GOF
a
Two molecules per asymmetric unit. A diffusely formed and
positionally disordered small molecule remote from either compound
was treated and refined as four carbon atoms per asymmetric unit,
b
c
hence the formula weight given. R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 =
d
[∑w(|Fo| − |Fc|)2/∑wFo ]1/2. GOF (all data) = [∑w(|Fo| − |Fc|)2/(n
− p)]1/2; n = number of independent reflections, p = number of
parameters.
2
3, important metric parameters are given in Table 4, and a
molecular view is presented in Figure 2. The tungsten methyl
derivative is a pseudotetrahedral molecule with N−W−C angles
that average 103.6(7)° and N−W−N angles that average
114.7(10)°, reflecting the steric requirements of the ligands.
Bis-imido-amido derivatives of tungsten are prone to a 3-fold
disorder averaging the two bulky ligands, and 1-Me is no
exception, as the three d(WN) values average 1.825(11) Å,32,33
with W−N−Si angles that average 160.4(8)°. The d(WC) value
is 2.158(6) Å, consistent with a standard sp3 tungsten−carbon
bond.
Figure 2. Molecular view of (tBu3SiNH)(tBu3SiN)2WCH3 (1-
CH3); note that imido and amido groups are disordered.
WCH2CH3 (1-CH2CH3) in 30% yield, accompanied by the
solvent-activated product (tBu3SiND)(tBu3SiN)2WC6D5 (1-
ND-C6D5, 23%), the KI adduct [(tBu3SiN)3WI]K (2-I-K,
27%), and the neutral iodide (tBu3SiN)2(tBu3SiNH)WI (1-I,
21%). When CD3CH2I was used, the amount of 1-I remained
constant at 21%, suggesting it was formed by a path, possibly a
radical or electron transfer process, different from the one
generating the remaining products: i.e., the hydride delivery
path. The decline in ethane-activated products with respect to
KI adduct upon deuteration is reminiscent of the MeI case
discussed above. Both CH- and CD-activated ethane products
(17%) are observed, with 1-CH2CD3 dominating over 1-ND-
[(tBu3SiN)3WH]M (2-H-M) and Saturated Alkyl
Halides. 1. Ethyl Halides. Utilization of deuterated EtX
substrates helped determine whether the site of activation was
the same as the site of hydride delivery (Scheme 5). Treatment
of [(tBu3SiN)3WH]K (2-H-K) with EtI at 100 °C in C6D6
afforded the ethyl complex (tBu3SiN)2(tBu3SiNH)-
6524
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 5
CD2CH3 in a 15.1:1.9 ratio, implicating a kH/kD value of 7.9(3)
for activation from an ethane complex intermediate,
(tBu3SiN)3W(CH3CD3) (2-CH3CD3). No scrambling be-
tween α and β sites of the W-CH2CD3 and W-CD2CH3 groups
was detected by NMR spectroscopy; thus, it is clear that the
hydride was transferred to the CH2 group of the substrate, and
no β-elimination processes that lead to scrambling were
evident. The decline in CD3CH3-activated products relative
to 1-ND-C6D5 may be an indication that deuteration weakens
the binding in the putative alkane complex intermediates. It is
likely the binding of ethane in the (tBu3SiN)3W pocket is
asymmetric, i.e., fast and reversible (tBu3SiN)3W···CH3CD3
↔ (tBu3SiN)3W···CD3CH3; hence, the KIE should be
considered a phenomenological value encompassing both
preferential binding and activation events.7 In this sense, if
(2)
ethane is more readily lost from (tBu3SiN)3W···CD3CH3, the
KIE could be even more skewed in favor of CH activation due
to an induced kinetic isotope effect.34
Unlike the methyl cases, the switch from alkyl iodide to
CH3CH2Br (trial 5.3) afforded a significant increase in the
amount of activation products40% 1-CH2CH3 and 31% 1-
ND-C6D5relative to 2-Br-K and 1-Br. The diminished
amount of 1-Br still fits with its generation via a separate
Scheme 6
6525
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 7
path, and its decline may be the origin of the greater amount of
other products. Note that the ratio of 1-CH2CH3:1-ND-C6D5 is
essentially independent of X, perhaps another indication that
solvent activation derives from RH loss from (tBu3SiN)3W-
(RH) (2-RH). It is difficult to spot a correlation in the
production of 1-X to other species, hence more alkyl halides
were scrutinized.
question by trials 6.1 and 6.2, since the 1-nPr/1-ND-C6D5 ratio
is dependent on substrate. However, this is the first case that
differential alkane binding is not caused by deuterationin this
instance primary CH vs secondary CHand loss of propane
from different sites may play a role. It is plausible that propane
is more easily lost from a secondary binding site, i.e., W(η2-
H2C(CH3)2, and this is the initial binding mode generated
upon hydride delivery; provided propane loss is competitive
with migration to the methyl termini, the scenario of a common
set of intermediates could still be satisfied.
The addition of EtI to 2−H-K in THF-d8 at 23 °C provided
100% [(tBu3SiN)3WI]K (2−I-K) according to eq 2. It
appears that the combination of a larger substrate and donor
solvent drives the reaction toward KI adduct formation.
2. Propyl Halides. Higher temperatures were necessary for
reasonable rates when propyl halide substrates were explored in
reactions with [(tBu3SiN)3WH]K (2-H-K), as shown in
Scheme 6. With nPrBr, 75% activation products were produced,
with the distribution roughly equal between
(tBu3SiN)2(tBu3SiNH)WnPr (1-nPr, 35%) and C6D6-activa-
tion product (tBu3SiND)(tBu3SiN)2WC6D5 (1-ND-C6D5,
40%). The remaining 25% was [(tBu3SiN)3WBr]K (2-Br-K),
and no detectable 1-Br was noted. No secondary activation of
3. Secondary, Tertiary, and Hindered Saturated Halides.
i
Given the results with PrX, other secondary halides, tertiary
halides, and hindered primary substrates were examined to see
if any (tBu3SiN)2(tBu3SiNH)WR (1-R) could be observed
upon exposure to [(tBu3SiN)3WH]K (2-H-K). Scheme 7
reveals that for isobutyl bromide and cyclohexylmethyl
bromide, no tungsten alkyl product was observed, but the
amounts of (tBu3SiND)(tBu3SiN)2WC6D5 (1-ND-C6D5)
and [(tBu3SiN)3WBr]K (2-Br-K) were greater than that of
(tBu3SiN)2(tBu3SiNH)WBr (1-Br), a result accommodated
by either hydride delivery or radical chain paths. A switch to
neoPeX substrates again gave no 1-neoPe, but more 1-X was
produced for X = I vs X = Br, and use of 2-H-Na and neoPeCl
afforded only 2-Cl-Na. tBuX substrates provided mostly 1-X, no
1-R, and a lesser amount of 1-ND-C6D5, consistent with a shift
to a radical path, replete with isobutene and isobutylene as
organic products. As steric factors are increased, hydride
delivery appears to become the minor path and radical
mechanisms start to dominate.
4. Radical Probes. Given the switch to potential radical-
based mechanism(s) as steric factors increase, and as secondary
and tertiary substrates were introduced, two radical probes were
investigated.35−37 The rate constant for the rearrangement of
cyclopropylmethyl radical to 3-butenyl radical is 8.6 × 107 s−1 at
25 °C;35,36 hence, cyclopropylmethyl bromide was chosen as a
substrate with the potential to ferret out radical character. As
i
i
propane was indicated, and use of PrBr and PrI also failed to
elicit an isopropyl tungsten compound. Instead, isopropyl
bromide generated a small amount of 1-nPr (2%), about 10% of
the solvent-activated complex 1-ND-C6D5, and 72% of the
major product, the KBr adduct 2-Br-K. In addition, 16% of
(tBu3SiN)2(tBu3SiNH)WBr (1-Br) was observed, and this
increased to 38% when isopropyl iodide was the substrate. With
the latter, no 1-nPr was produced and the remainder of the
material was 2-I-K (56%) and 1-ND-C6D5 (6%). Accompany-
ing 1-Br were propane, propene, and 2,3-dimethylbutane,
products that are suggestive of a radical process, since they can
i
be derived from Pr radical. In THF-d8, the combination of 2-
i
H-K and PrI afforded 100% 2-I-K, akin to the related
experiment with EtI (eq 2).
The premise that 1-R and 1-ND-C6D5 originate from a
commonor set of commonintermediates is called into
6526
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 8
Scheme 9
Scheme 8 illustrates, a 10-fold excess of substrate with 2-H-K in
C6D6 for 6 h at 100 °C afforded mostly (84%) the
nonrearranged alkyl product (tBu3SiN)2(tBu3SiNH)-
W-cPrMe (1-cPrMe), and the activation occurred solely at a
nonmethylated secondary cyclopropyl site rather than the
methyl. Multidimensional NMR correlation techniques pos-
itively identified the conformation, which was ultimately
confirmed by an X-ray crystallographic study, albeit one with
poor-quality data. A minor product (4%) in the process was the
rearrangement product (tBu3SiN)2(tBu3SiNH)W(trans
-CHCHEt) (1-CHCHEt), but 1-Br (2%) was also
formed; hence, the possibility exists that 1-CHCHEt was
derived solely from a separate radical path.
time scale roughly on par with that of rearrangement at 100 °C
(∼1010 s−1).35,36 Unfortunately, while rearrangement of cyclo-
propylmethyl radical is fast, so is the rearrangement of
cyclopropylmethyl cation; hence, it seems reasonable that the
Lewis acidity of the tungsten center could also effect some
rearrangement. Given the predominance of unrearranged
product, dominance of a hydride delivery path is certainly
reasonable. Note that both alkane complex intermediates,
unrearranged 2-HcPrMe and rearranged 2-1-butene, are
activated at sp2 carbons, at a ring site anti to the methyl in
the former and at the trans-alkene terminus in the latter,
thereby reflecting selectivity patterns previously observed in
early-metal imido 1,2-RH additions.20,23,38
1
Upon scale-up of the reaction in benzene (trial 8.2), H
A second radical probe, albeit a significantly slower one, is
the 5-hexenyl radical, which rearranges to the cyclopentyl-
methyl radical with a unimolecular rate of 2.0 × 105 s−1 at 25
°C.35,37 No rearrangement product is observed in this instance,
consistent with the rate discussion above, and the tungsten sp2-
activated alkyl (tBu3SiN)2(tBu3SiNH)W(trans-CH
NMR analysis of the crude reaction mixture showed no 1-Br or
2-Br-K, yet 10% rearrangement occurred along with 90%
unrearranged 1-cPrMe (52% isolated yield). Since the lack of an
observed amount of 1-Br does not completely rule out radical
paths, any such path would require hydride abstraction on a
6527
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
CHnBu) (1-CHCHnBu) was identified in the crude reaction
mixture in 80% yield. The remaining constituents are 2-Br-K
(9%) and 1-Br (11%). Since (tBu3SiN)2(tBu3SiNH)-
c
c
WCH2 Pe (1-CH2 Pe) was previously shown to be stable to
the reaction conditions,30 the most likely scenario for the
production of 1-CHCHnBu is illustrated in Scheme 9.
Hydride delivery by 2-H-K affords a 1-hexene adduct, 2-1-
hexene, that is activated at its sp2 carbon end after migration
down the alkyl chain.20,23,38
(3)
5. Structure of (tBu3SiN)2(tBu3SiNH)W-cPrMe (1-cPrMe).
An X-ray crystallographic study of (tBu3SiN)2(tBu3SiNH)-
W-cPrMe (1-cPrMe) was undertaken, and details of the
crystallographic and data refinement can be found in Table 3,
while metric parameters are given in Table 4. It was a poor data
set, and an isotropic model was the most appropriate solution
to the refinement. One of the two independent molecules in
the asymmetric unit was rife with disorder in the peripheral tBu
groups; thus, Figure 3 depicts the least problematic structure.
spectrum consistent with the formulation [(tBu3SiN)3WPh]
K (2-K-Ph, 22%, δ 1.43 (s, 81 H), 6.59 (t, 8 Hz, p-H), 6.80 (t, 8
Hz, m-H), 8.40 (d, 8 Hz, o-H)) (eq 4). The spectral
(4)
characteristics were similar to those of [(tBu3SiN)3WPh]Li
(2-Li-Ph), but without independent synthesis, assignment of
this minor product remains tentative. The most likely arene
activation products, (tBu3SiNH)(tBu3SiN)2WPh (1-Ph) and
1-ND-C6D5, were not discerned in the product mixture.
Previous studies in d0 RH activation have shown a selectivity
for sp2 over sp3 CH bonds,20,23,38 a factor that presumably
determined the 1,2-RH addition of a cyclopropyl CH rather
than a methyl CH in Scheme 8 and a vinylic CH rather than an
aliphatic CH in Scheme 9. When [(tBu3SiN)3WH]K (2-H-
K) was treated with vinyl bromide, a reaction occurred at a
reasonable rate only at high temperatures (140 °C) over several
days, producing a multitude of unidentified products. Even 2-
Br-K and 1-Br were not positively identified. Vinyl halides have
reaction paths such as insertion30 and polymerization that are
not available to other substrates, and these may supersede
hydride delivery paths, producing complexes that have not been
prepared by other means.
2. Benzylic and Allylic Halides. A smattering of benzylic
substrates were evaluated, and consistent with previously noted
selectivities,20,23,38 aryl CH bond activation predominated.39,40
As Scheme 10 illustrates, exposure of [(tBu3SiN)3WH]K (2-
H-K) with BnI at 23 °C for 13 h in benzene-d6 afforded two
aryl activation products, (tBu3SiNH)(tBu3SiN)2W(C6H4-p-
Me) (1-p-tol, 53%) and 38% (tBu3SiNH)(tBu3SiN)2W-
(C6H4-m-Me) (1-m-tol, 38%), along with 9% of
[(tBu3SiN)3WI]K (2-I-K) and trace toluene. No activation
of the benzylic position was observed, consistent with the
known preference for aryl over benzyl 1,2-RH addition. The
same reaction when carried out in toluene-d8 yielded essentially
the same ratio of toluene-activated products but an increased
amount of 2-I-K (21%); most importantly, no (<3%) activation
of C7D8 was noted. A switch to BnCl as the substrate and a
change in conditions (60 °C, 11 h) provided a 46/41 ratio of
para to meta activation products and 13% 2-Cl-K.
Figure 3. Molecular view of the less disordered of two independent
molecules of (tBu3SiN)2(tBu3SiNH)W-cPrMe (1-cPrMe); data
quality led to an isotropic refinement.
Although the thermal parameters of its methylcyclopropyl
carbons were large, the model was good enough to confirm that
the Me and W were disposed anti about the cyclopropane ring,
as shown by the multidimensional NMR studies. Despite the
crystallographic problems, the core values were normal, with
the usual imide and amide disorder: d(WN)av = 1.824(13) Å,
d(W−C)av = 2.16(8) Å; ∠NWNav = 113.2(17)°, ∠CWNav
=
105(4)°.
[(tBu3SiN)3WH]M (2-H-M) and Unsaturated Sub-
strates. 1. sp2-Carbon Halides. Since (tBu3SiND)-
(tBu3SiN)2WC6D5 (1-ND-C6D5) was a typical byproduct
in many of the studies discussed, PhX was a logical substrate.
Treatment of [(tBu3SiN)3WH]K (2-H-K) with PhBr at 150
t
°C for 24 h produced 51% (tBu3SiN)2( Bu3SiNH)WBr (1-
Br), 24% [(tBu3SiN)3WBr]K (2-Br-K), and unidentified
tungsten products (eq 3). A switch to PhI enabled reduction to
occur at lower temperatures, and [(tBu3SiN)3WI]K (2-I-K,
12%) and (tBu3SiN)2(tBu3SiNH)WI (1-I, 51%) were
In the case of xylyl bromide (trial 10.4), benzylic activation
was observed, as treatment of 2-H-K with BrCH2C6H4-p-CH3
at 23 °C for 20 h afforded 14% (tBu3SiNH)(tBu3SiN)2W-
(CH2C6H4-p-Me) (1-xyl) in addition to the major (86%) aryl-
activated product (tBu3SiNH)(tBu3SiN)2W(C6H3-2,5-Me2)
1
generated along with a set of resonances in the H NMR
6528
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 10
Scheme 11
Scheme 12
(1-2,5-Me2Ph). No solvent, i.e., benzene-d6, activation or halide
adduct formation was seen; hence, this case is an unusual
example where the products are solely due to CH bond
activation. A similar result was obtained for apparent hydride
6529
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
transfer from 2-H-K to 2-bromomethylnaphthalene at 60 °C
for 5 h in C6D6 (trial 10.5), where only the 2-naphthalenyl
derivatives (tBu3SiNH)(tBu3SiN)2W(2-C10H6-6-Me) (1-6-
Me-2-nap) and (tBu3SiNH)(tBu3SiN)2W(2-C10H6-7-Me)
(1-7-Me-2-nap) are formed in a ∼16/9 ratio. In the case of
the hindered benzylic substrate mesityl bromide, solvent
activationin this case formation of 20% (tBu3SiND)-
(tBu3SiN)2WC6D5 (1-ND-C6D5) from C6D6reemerged
as a component in the reaction, and 68% of the product
mixture was due to benzylic activation of mesitylene, as
evidenced by identification of ( t Bu3 SiNH)-
(tBu3SiN)2WCH2C6H4-3,5-Me2) (1-mes). The halide adduct
2-Br-K was formed in 12% yield, and a trace of mesitylene was
noted.
resonances at δ 2.62 (d, 2 Hz, 3H) and δ 5.89 (q, 2 Hz, 1H),
and a large JWH value of 54 Hz accompanying the latter, were
indicative of a trans three-bond linkage between tungsten and
the proton.
3
Competition between CH activation and cycloaddition has
been previously observed in titanium19,43 and related vanadium
systems, and the latter underwent ring expansion reactions.25,26
When bromo-2-butyne was the substrate, the SN2′ path was not
observed, presumably due to steric hindrance of hydride
delivery to methyl-containing alkyne carbon, and only the
tungstacycle (tBu3SiN)2W(κ-N,C-N(SitBu3)C(Me)C(Me)
−) (3-MeCCMe, 64%) and a rearrangement product (11%)
were found after 24 h at 23 °C (C6D6) in addition to 2-Br-K
(25%). The 1H NMR spectrum of 3-MeCCMe resulted in one
sharp resonance at δ 1.33 (54 H) and one broad resonance at δ
1.21 (27 H), which were assigned to the free and cyclized imido
ligands, in addition to those of the minor rearrangement
product. When the temperature was raised to 70 °C, the δ 1.21
resonance sharpened, consistent with restricted rotation of the
tBu groups in the metallacycle. Distinct Me resonances (δ 2.28
In Scheme 11, the treatment of [(tBu3SiN)3WH]K (2-H-
K) with H2CCHCH2X is shown, and both the E and Z
isomers of propenyl activation, (tBu3SiNH)(tBu3SiN)2W(E-
CHCHMe) (1-E-CHCHMe) and (tBu3SiNH)-
(tBu3SiN)2W(Z-CHCHMe) (1-Z-CHCHMe) are the
products, with no other species formed when X = I. For X = Cl,
∼10% 2-Cl-K is also generated. It is not clear whether any 1-Z-
CHCHMe is derived directly from activation, as the E/Z
ratio was found to be dependent on the amount of exposure to
room light. Subsequent studies revealed that the cis isomer is
produced in about 45% yield at 23 °C in a photostationary state
under ambient conditions, and extended thermolysis of 1-E-
CHCHMe in the dark failed to generate any 1-Z-CH
CHMe. Once again, sp2 activation was preferred over allylic
activation;20,23,38 hence, the plausible propene adducts of tris-
imido tungsten appear to activate from a single conformation.
The examination of propargylic substrates provided evidence
of diversion from the standard reactivity of the previous
substrates. As Scheme 12 shows, treatment of
[(tBu3SiN)3WH]K (2-H-K) with 3-bromopropyne at 60
°C for 24 h in C6D6 gives 13% of the halide adduct
[(tBu3SiN)3WBr]K (2-Br-K), a small amount of
(tBu3SiN)2(tBu3SiNH)WBr (1-Br, 4%), and 58% of the
allene-activated complex (tBu3SiN)2(tBu3SiNH)WCH
CCH2 (1-CHCCH2). The last species is consistent
with an SN2′ delivery of hydride by 2-H-K to provide an
intermediate allene complex, (tBu3SiN)3W(allene) (2-
H2CCCH2), from which CH activation occurs. Since 1-
CHCCH2 was the major product, the reaction was scaled
up and the allenyl species41,42 was isolated as a yellow powder
in 31% yield. A strong, sharp infrared band at 1915 cm−1 was
assigned to the asymmetric CCC mode common to allenic
units, and 1H NMR resonances (C6D6, Table 1) at δ 6.70 (t, 7
1
and δ 2.52 (3JWH = 13 Hz)) were observed by H NMR
spectroscopy, and two methyl (δ 15.81 (3JWC = 26 Hz), δ 20.71
(2JWC = 9 Hz)) and two alkenyl (δ 126.23 (2JWC = 15 Hz), δ
174.24 (1JWH = 161 Hz)) carbon resonances were observed by
13C{1H} NMR spectroscopy. These did not change upon
warming; thus, reversible formation of the metallacycle is not
facile. No attempts to separate 3-MeCCMe from its rearrange-
ment product were attempted.
Thermolysis of 3-MeCCMe for 6 h at 63 °C caused
complete conversion to the rearrangement product, the
cyclometalated azaallyl (tBu3SiN)2W(κ-C,η3-C,N-
CH2CMe2Si(tBu2)NCMeCHMe) (4-MeCCMe), analogous to
similar rearrangements derived from 2-butyne and 3-hexyne
additions to (tBu3SiN)2(tBu3SiNH)V(OEt2).25,26 NMR
spectroscopic characterization of azaallyl 4-MeCCMe revealed
t
inequivalent SitBu3, Bu, CMeMe, and CHH groups, consistent
with the asymmetry derived from the facial binding of the
azaallyl fragment (Table 1). In the 13C{1H} NMR spectrum,
tungsten couplings to the azaallyl carbons at δ 53.28 (NCC, JWC
= 24 Hz) and δ 180.47 (NC, JWC = 10 Hz) were modest in
comparison to coupling with the metallacyclic methylene (δ
54.08 (JWC = 105 Hz)), consistent with the π character of the
tungsten-azaallyl bonding. While 4-MeCCMe was only
characterized by NMR spectroscopy, its strong spectral parallels
to the structurally characterized vanadium example lend
credence to the posited geometry shown in Scheme 12.
3. Structure of (tBu3SiN)2(tBu3SiNH)WCHCCH2 (1-
CHCCH2). An X-ray crystal structure determination of
(tBu3SiN)2(tBu3SiNH)WCHCCH2 (1-CHCCH2)
was conducted, and a view of the molecule is displayed in
Figure 4, while data collection and refinement parameters are
given in Table 3 and selected bond distances and angles are
given in Table 4. The disorder typical for the
(tBu3SiN)2(tBu3SiNH)W fragment is less evident in 1-
CHCCH2 as one W−N distance is statistically shorter
(1.782(4) Å) than the other two (1.850(4), 1.860(4) Å). The
short distance is reasonable for a tungsten imide,32,33,44 and the
remaining two distances may reflect an imide/amide average.
The W−N−Si angle for the imide (162.6(3)°) is greater than
that of the others (157.3(2), 157.4(3)°), but these angles are
known to not necessarily correlate with an imide vs an amide.
The core N−W−N and N−W−C angles average 114.3(28)
and 104.1(19)°, respectively. The d(WC) value of 2.155(6) Å
2
4
Hz, JWH = 7 Hz, 1H) and δ 4.35 (d, 7 Hz, JWH = 7 Hz, 2H)
were assigned to the CH and CH2 hydrogens. The 13C{1H}
NMR spectrum was also diagnostic, with a resonance at δ
1
109.08 accompanied by JWC = 179 Hz indicative of the α-
carbon and resonances at δ 212.07 and δ 67.08 (3JWC = 8 Hz)
consistent with the sp and sp2 carbons, respectively. Spectral
characteristics for 1-CHCCH2 closely matched those of
Cp(CO)3WCHCCH2,41 although its JWC value of 47 Hz
1
is attenuated due to its higher coordination number and
correspondingly less tungsten s character in the W−C(α) bond.
In the reaction with propargyl bromide, SN2 hydride delivery
was expected to generate an intermediate propyne adduct,
(tBu3SiN)3W(H3CCCH) (2-HCCCH3), but instead of sp
CH activation, 2π + 2π cycloaddition occurred to provide a
tungstaazacyclobutene, (tBu3SiN)2W(κ-N,C-N(SitBu3)CHC-
(Me)−) (3-HCCMe) as the remaining product. Coupled
6530
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
produced upon breakup of a single intermediate, I. While this
mechanism cannot be disproved, the various trials in Schemes 2
and 3 show that variation in solvent at 60 °C, e.g., Et2O vs
C6D6, with CD3I as substrate does not change the ratio of
(tBu3SiN)2(tBu3SiN(H/D))W(CD3/CHD2) (1-CD3/1-ND-
CHD2) to 1-ND-C6D5 or 2-OEt2. If the solvent were intimately
involved in an intermediate such as I, significant deviations
would be expected.
The ratio {1-Me + 1-ND-C6D5}/2-I-K does change as a
function of solvent (Schemes 2−4) but does not change as X is
varied in CH3X. The former is consistent with intermediate I in
Scheme 13 but is generally consistent with alternative
processes, while the latter suggests that the transition state
regarding reduction, either a hydride delivery or eT step, is
early in a reaction coordinate whose Δ G° value is likely to be
substantially favorable. Hydride delivery is implicated by
intermediate I, yet it is conceivable that electron transfer to
give products within a radical cage is also reasonable, provided
the lifetime of R• is short compared to the time scale of
rearrangement of radicals derived from radical clocks cPrCH2Br
and 5-hexenyl bromide. Note that cage collapse of either rc1 or
rc2 in Scheme 13 to intermediate I or to products 1-R or RH
via hydrogen atom abstraction (HAT)45,46 provides a plausible
alternative to hydride transfer, albeit on the appropriate time
scale.
Scheme 14 illlustrates the double intermediate mechanism used
in describing the majority of the product distributions in
Schemes 2−12. In this mechanism, the putative intermediate I
degrades into [(tBu3SiN)3WX]M (2-X-M) via loss of RH or
an alkane complex intermediate, (tBu3SiN)3W(RH) (2-RH),
a n d M X . C o n v e r s i o n o f t h e l a t t e r i n t o
( t B u 3 S i N  ) 2 ( t B u 3 S i N H ) W R ( 1 - R ) a n d
(tBu3SiN)2(tBu3SiND)WC6D5 (1-ND-C6D5, solvent =
C6D6) or (tBu3SiN)3WL (L = donor solvent) upon loss of
RH completes the scheme. It is also possible that I is a
transition state that partitions to 2-X-M and 2-RH. Variants of
intermediate I based on eT processes shown in Scheme 13 can
also apply. Intermediate 2-RH is likely to swiftly equilibrate
with other bound forms of the alkane, i.e., (tBu3SiN)3W-
(R′H) (2-R′H), such that previously established CH bond
selectivities apply.20,23,38 The alkane complex intermediate is
supported by calculations,15−18 is fully consistent with 1,2-RH
addition reactions observed for other imido compounds,20−28,38
fits the observed changes in product ratios, etc.
Figure 4. Molecular view of (tBu3SiN)2(tBu3SiNH)WCHC
CH2 (1-CHCCH2).
is roughly the same as that of 1-CH3 but slightly shorter than
that of the methylcyclopropyl derivative, and the allene unit is
almost linear (∠CCC = 178.0(10)°), with CHC and C
CH2 distances of 1.183(8) and 1.320(11) Å, respectively.
Trimethylsilyl Bromide. The question of whether the
“alkane intermediate” approach could be extended to other
elements was only broached through the addition of Me3SiBr to
[(tBu3SiN)3WH]K (2-H-K). In less than 1 h in C6D6, 95%
(tBu3SiN)2(tBu3SiNH)WSiMe3 (1-SiMe3) was produced
with 5% (tBu3SiN)2(tBu3SiNH)WBr (1-Br) as a byproduct,
as eq 5 indicates. Hydride delivery to the substrate, followed by
(5)
SiH activation from the putative (tBu3SiN)3W(HSiMe3) (2-
HSiMe3) intermediate, is a logical path to 1-SiMe3.
Kinetic Isotope Effects. The phenomenological KIEs
determined for [(tBu3SiN)3WH]K (2-H-K) plus CD3I can
be used to test the consistency of the double intermediate
DISCUSSION
■
General Mechanism. Illustrations accompanying the
preceding results sections have been predicated on the
generation of hydrocarbon complex intermediates in the
reactions of [(tBu3SiN)3WH]M (2-H-M) with hydrocarbyl
halides. Extrapolating from the modest kinetics treatment of 2-
H-K and MeX, the general reaction is best considered first
order in 2-H-M and first order in RX (Scheme 4, Table 2). The
reduction step (RX + MH → RH + MX) can be construed as
possessing hydride or electron transfer (eT) character, as the
rates of MeX reduction (k(I−) ≈ k(Br−) > k(Cl−)) are not
particularly informative.
Scheme 13 illustrates the single intermediate mechanism
w h e r e b y t h e t h r e e s t a n d a r d p r o d u c t s ,
(tBu3SiN)2(tBu3SiNH)WR (1-R), [(tBu3SiN)3WX]M (2-
X - M ) , a n d t h o s e d e r i v e d f r o m s o l v e n t ,
(tBu3SiN)2(tBu3SiND)WC6D5 (1-ND-C6D5, solvent =
C6D6) or (tBu3SiN)3WL (2-L, L = OEt2, THF, etc.), are
mechanism via scrutiny of the [1-Me]/[1-ND-C6D ] ratio vs
5
the ([1-CD3] + [1-ND-CHD2])/[ 1-ND-C6D5] ratio, as
Scheme 15 indicates.34 In the analysis, factors affecting the
ratios have been kept simple. For example, since the structure
of (tBu3SiN)3W(CHD3) (2-CHD3) is unknown, i.e. η-
CHD3, η2-CHD3, or η3-CHD3, it is treated as a single entity,
because any further complication could not be untangled with
the limited data at hand.7 Only the CD3 isotopologue of methyl
iodide was examined; hence, secondary isotope effects cannot
be teased out of the data. The analysis thus uses two KIEs: the
phenomenological kH/ kD (za) accorded activation of the CH of
bound CHD3 relative to 1,2-CD-addition21,24,28 and a zs
pertaining to the dissociation of CH4 vs CHD3, a value that
7,49
reflects their relative binding.
As Scheme 15 indicates, while the za value of 9.6 is sufficient
to explain the change in ratios upon deuteration with respect to
6531
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 13
Scheme 14
experimental error, solving for zs affords a value <1, consistent
with weaker binding of CHD3. Weaker binding of deuterated
methanes is due to greater attenuation of the CD bond strength
when deuterium is in a bridging position: i.e., W(μ-D)C. The
subtly greater weakening of a CD bond vs the corresponding
CH bond of bound CH4 is enough to bias the binding such that
CHD3 is more readily lost: hence, the zs value of 0.94 obtained
from the analysis.7,49 The change in ratios leading toward more
solvent activation in the deuterated case is thereby ascribed to
two factors: (1) 1,2-CD addition is slower by a factor of 9.6,
hence, loss of methane becomes more competitive with
activation and (2) loss of CHD3 from 2-CHD3 occurs slightly
more readily than CH4 from 2-CH4,7 leading to slightly more
solvent activation.34
Since the data acquired in the investigation of methane
activation are limited, corroboration of the KIE arguments
above was sought for the case of CH3CH2I vs CD3CH2I, and
this is illustrated in Scheme 16. In this analysis, the assumptions
delineated above were again applied, with a single KIE that
describes 1,2-CH vs 1,2-CD addition to the imide linkage (kH/
kD = za) to form (tBu3SiN)2(tBu3SiNH)WCH2CD3 (1-
CH2CD3) and (tBu3SiN)2(tBu3SiND)WCD2CH3 (1-ND-
CD2CH3) and a single KIE describing loss of ethane from
(tBu3SiN)3W(CH3CH3) (2-CH3CH3) vs (tBu3SiN)3W-
(CH3CD3) (2-CH3CD3) (kH/kD = zs). The ratio of ratios was
obtained from the product breakdown in trials 5.1 and 5.2, and
a za value of 7.9 yielded a zs value <1. Again, the change in
ratios can be explained solely by the slowdown in 1,2-CD
addition from 2-CH3CD3 within the constraints of exper-
imental error, but the calculated addition of zs = 0.89 is
consistent with weaker binding of CH3CD3 as a contributing
factor.7
Radical Paths. Since the two radical clocks employed in
this study failed to elicit evidence of radical paths, methyl and
primary halides are considered reduced by hydride transfer or
by radical processes swift enough to bypass the clock
rearrangements. In the cases of secondary and tertiary halides
and ethyl, evidence for radical processes exists in the product
distributions.47,48 Consider the standard radical processes
summarized in Scheme 17, where the initial electron transfer
(I1) from [(tBu3SiN)3WH]M (2-H-M) plus RX yields R•
and a metallaradical. A subsequent halide abstraction by the
6532
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 15
Scheme 16
metallaradical (I2) can generate another 1 equiv of R• and
(tBu3SiN)2(tBu3SiNH)WX (1-X).
For the tert-butyl cases, 0−2% of 2-X-K is produced, and the
∼1/1 ratio of isobutylene and isobutene products suggests that
propagation steps (P1, P2) leading to 2-X-K are not
competitive. In contrast, the less hindered neopentyl radical
appears likely to abstract a hydrogen atom from 2-H-K in a
propagation sequence, leading to substantial amounts of 2-X-K,
and in the case of neoPeCl and 2-H-Na (trial 7.3), the 100%
formation of 2-Cl-Na is consistent with an efficient radical
Scheme 7 shows that 1-X is a prominent product for tert-
butyl and neopentyl halides, and for the latter substrate it
increases in the order X = Cl < Br < I, consistent with halide
abstraction. Even in the cases of EtBr and EtI, significant
amounts of 1-X (X = Br, I) are generated. For the hindered
primary, secondary, and tertiary substrates in Scheme 7, the
amounts of 2-X-K and 1-X encompass ≥69% of the products.
i
chain. Likewise, in the case of PrI reduction, a greater amount
6533
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
Scheme 17
of 2-I-K than of 1-I was produced, and all three organic
termination products expected for isopropyl radicalpropane,
propene, and 2,3-dimethylbutaneare observed. In the
isobutyl (trial 7.1) and cyclohexylmethyl bromide (trial 7.2)
substrates, the amount of RH formed is greater than that of the
unsaturated organics; thus, chain processes may account for the
majority of RH. While it was difficult to determine whether the
product ratios were constant or changing during the course of
reduction,48 there is enough evidence to conclude that radical
processes are competent for substrates that are hindered. The
situation is very similar to observations of oxidative additions of
RX to reduced metal centers, where SN2 vs radical mechanisms
are extremely sensitive to the size and type of R.
Selectivities. Selectivities for the type of CH bond
activated via the intermediate alkane (arene) species
(tBu3SiN)3W(RH) (2-RH) follow previously established
trends.20,23,38−40 For example, in instances where sp3-CH vs
sp2-CH bond activations are competing, the product was
invariably derived from the latter. In the related cases of 2-RH
(RH = toluene, xylene, mesitylene),39,40 the product ratios at
23 °C could be used to assess the relative ΔΔG⧧ value for arene
activation at sites with no o-Me (A0), one o-Me (A1), and two
o-Me (A2) groups versus benzylic activation (B) and arene loss
(E). On comparison of the product ratios in Scheme 10 and
correction for statistics, the following relative activation order
was ascertained: A0 (both para and meta activation of toluene,
i.e., [1-p-tol] and [1-m-tol]) is lowest; A1, ΔΔG⧧ > 2.5 kcal/
mol relative to A0; B, ΔΔG⧧ > 2.7 relative to A0, ΔΔG⧧ = 1.3
kcal/mol relative to A1; E, ΔΔG⧧ = 0.75 kcal/mol relative to B;
A2, ΔΔG⧧ > 2.1 kcal/mol relative to B. Activation of toluene at
the para and meta positions is roughly favored over ortho
activation by >2.5 kcal/mol and over benzyl activation by >3.8
kcal/mol, while arene activation between two methyl groups is
disfavored by >5.9 kcal/mol. The relative selectivities appear
general to oxidative addition reactions39,40 and 1,2-RH
additions to imido units.
CONCLUSIONS
■
In cases of unhindered sp3 alkyl halides, reduction of RX by
[(tBu3SiN)3WH]M (2-H-M) proceeds to afford
[(tBu3SiN)3WX]M (2-X-M) and RH or loss of MX and
products derived from an alkane or arene complex
intermediate, (tBu3SiN)3W(RH) (2-RH). For hindered sp3
substrates, radical paths change the nature of the product
distribution toward halide products. Through exhaustive studies
of 2-H-M plus RX, a consistent pattern has emerged that
dovetails with previous studies regarding the 1,2-RH elimi-
nation from XY(tBu3SiNH)MR and 1,2-RH addition to
transient XY(tBu3SiN)M.
The indirect observation of various (tBu3SiN)3W(RH) (2-
RH) suggests that intermediates (tBu3SiO)2(tBu3SiN)Ti-
(RH), 2 0 , 2 1 (t Bu3 SiNH)2 (t Bu3 SiN)Ti(RH), 2 2
6534
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
(t Bu3 SiNH)2(t Bu3SiN)Zr(RH),23,24 (t Bu3SiNH)
(tBu3SiN)2V(RH),25,26 and (tBu3SiNH)(tBu3SiN)2Ta-
(RH)27 are likely to be operable in these related systems. In
low-valent late-transition-metal species, LnM(RH) reactions
also precede C−H bond activation to give LnM(H)R, and
activation selectivities are determined by RH binding.12,39,49 In
contrast, 1,2-RH addition is the selectivity-determining event
for the titanium, zirconium, vanadium, and tantalum d0 imido
complexes. The standard free energy for 1,2-RH addition to
putative (tBu3SiN)3W (2) is very favorable; thus, it is
competitive with alkane loss in this investigation. Alkane loss
from 2-RH is noted by solvent activation to give
(tBu3SiN)2(tBu3SiND)WC6D5 (1-ND-C6D5, solvent =
C6D6). Phenomenological KIEs obtained via substrates CD3I
and CD3CH2I are in line with those obtained for CHD3 and
CH3CD3 activations by putative (tBu3SiO)2(tBu3SiN)Ti and
related imido complexes,21,24 providing additional evidence for
the intermediacy of 2-RH. In cases where 2-RH contains
different CH bonds, the selectivity of bond activation follows
previously established selectivity patterns;20,23,38 hence, it is
likely that a rapid equilibrium is established between various
binding conformations subsequent to generation of the initial
2-RH. On the basis of this work, previous KIEs determined for
1,2-RH/D addition to (tBu3SiO)2(tBu3SiN)Ti and related
imido species are likely to be comprised of binding and 1,2-
addition events.21,24
corresponded to the resonances of the CD2H of 1-ND-CD2H and
the CD3 of 1-CD3, while the resonance at δ 6.60 corresponds to the
ND of 1-ND-CD2H.
(b). C6H6 Solvent. 2-H-K (125 mg, 0.145 mmol) in 5 mL of
benzene and 469 Torr of CD3I (50.4 mL gas bulb, 1.28 mmol, 9
equiv) were stirred for 10 h at 60 °C. The products were assayed by
NMR spectroscopy (Scheme 2), and a 2H{1H} NMR spectrum
revealed the ratio of 1-ND-CD2H to 1-CD3 to be 3.2(2).
(c). Et2O Solvent. 2-H-K (200 mg, 243 mmol) in 10 mL of Et2O
and 185 Torr of CD3I (0.266 L gas bulb, 2.68 mmol, 11.5 equiv) were
stirred for 11 h at 60 °C. The products were assayed by NMR
2
spectroscopy (Scheme 3), and a H{1H} NMR spectrum revealed the
ratio of 1-ND-CD2H to 1-CD3 to be to be 3.3(2). A sample (10 mg)
of the crude reaction mixture was taken up in C6D6, sealed in an NMR
tube, and subjected to thermolysis at 150 °C. 2-OEt2 was shown to
convert to 1-ND-C6D5 with t1/2 < 10 h (assume unimolecular; ΔG⧧ ≈
34 kcal/mol).
(d). THF Solvent. 2-H-K (0.165 g, 0.191 mmol) in 10 mL of THF
and 132 Torr of CD3I (0.266 L gas bulb, 1.91 mmol, 10 equiv) were
stirred for 75 min at 100 °C. The products were assayed by NMR
2
spectroscopy (Scheme 3), and a H{1H} NMR spectrum revealed the
ratio of 1-ND-CD2H to 1-CD3 to be to be 3.9(2). A sample (10 mg)
of the crude reaction mixture was taken up in C6D6, sealed in an NMR
tube, and subjected to thermolysis at 150 °C. 2-THF slowly converted
to 1-ND-C6D5 with t1/2 < 10 h (assume unimolecular; ΔG⧧ ≈ 37 kcal/
mol).
2. (tBu3SiNH)(tBu3SiN)2WR (1-E/Z-CHCHMe). (a). Prepara-
tion. In a 100 mL glass bomb shielded from ambient light containing
2-H-K (640 mg, 0.740 mmol) was transferred 30 mL of benzene
followed by 512 Torr of allyl iodide by vacuum transfer (266 mL gas
bulb, 7.40 mmol, 10 equiv). The vessel was placed in a 60 °C bath for
8.5 h, and the initially colorless solution turned light yellow. After the
mixture was cooled to 23 °C and volatiles were removed, the
remaining solid was dissolved in 5 mL of hexanes, the solution was
filtered, and the solvent was replaced with THF. Subsequent cooling to
−78 °C gave a milky suspension which was collected as a light yellow
EXPERIMENTAL SECTION
■
General Considerations. All manipulations were performed
using either glovebox or high-vacuum-line techniques. All glassware
was oven-dried. THF and ether were distilled under nitrogen from
purple sodium benzophenone ketyl and vacuum-transferred from the
same prior to use. Hydrocarbon solvents were treated in the same
manner with the addition of 1−2 mL/L tetraglyme. Benzene-d6 and
toluene-d8 were dried over sodium, vacuum-transferred and stored
over activated 4 Å molecular sieves. THF-d8 was dried over sodium
and vacuum-transferred from sodium benzophenone ketyl prior to use.
All glassware was base-washed and oven-dried. NMR tubes for sealed-
tube experiments were flame-dried under dynamic vacuum prior to
use. Alkyl and aryl halides and other organic reagents were obtained
from Aldrich Chemical except for CD3CH2I (Cambridge Isotope
Laboratories) and 3,5-dimethylbenzyl bromide (Johnson Mathey).
Volatile alkyl halides were degassed, distilled, and dried over activated
4 Å molecular sieves, except for bromomethylcyclopropane, which
decomposed when exposed to sieves. Less volatile substrates were
subjected to degassing and used as received. The compounds
(tBu3SiNH)(tBu3SiN)2WX (1-X; X = Cl, Br, I), (tBu3SiNH)-
1
powder by filtration (180 mg, 28%). The H NMR spectrum of the
crude reaction mixture revealed the ratio of Z- to E-1-C(H)
C(H)Me to be 33(2). A corresponding NMR tube scale reaction
showed quantitative conversion of 2-H-K to tungsten vinyl products.
IR (Nujol, cm−1): 3249 (m, NH), 1578 (m, CC), 1468 (s), 1386
(s), 1364 (m), 1130 (s), 1055 (s), 1040 (s), 1009 (s), 971 (m), 934
(m), 846 (s), 819 (s), 643 (m), 621 (s), 582 (s). Anal. Calcd for
C39H87N3Si3W: C, 54.08; H, 10.12; N, 4.85. Found: C, 53.71; H, 9.99;
N, 4.67.
(b). Thermolysis and Photolysis. Samples of 1-CHCHMe were
loaded into two NMR tubes and dissolved in benzene-d6 and the tubes
sealed. One was shielded from light and heated to 150 °C for 7 days;
analysis by 1H NMR spectroscopy indicated no evidence of
isomerization. The second sample was subjected to a 20 W laboratory
fluorescent light for 12 h at 23 °C, affording a photostationary Z to E
ratio of 1.21(3)/1. Subsequent thermolysis in the dark at 150 °C for 7
days showed no deviation from this ratio.
n
(tBu3SiN)2WR (1-R; R = H, Me, Et, Pr, Ph, C6H4-3-Me, C6H4-4-
Me, Bn), [(tBu3SiN)3WH]M (2-H-M;
M = Na, K),
[(tBu3SiN)3WX]K (2-X-K; X = Cl, Br, I), and [(tBu3SiN)3WX]
Na (2-X-Na; X = Cl, Br) were either prepared or identified in an
accompanying paper.30
3. (tBu3SiNH)(tBu3SiN)2W(6/7-Me-2-nap) (1-6/7-Me-2-nap). In
a glass bomb reactor, 2-(bromomethyl)naphthalene and 2-H-K were
combined and 30 mL of benzene was added at −78 °C via distillation.
The mixture was heated to 60 °C for 11 h. After removal of volatiles,
the resulting solids were taken up in hexanes, and the solution was
filtered, concentrated, and cooled to −78 °C. 1-6/7-Me-2-nap was
collected as a light yellow solid by filtration (152 mg, 26%) and
assayed by 1H NMR spectroscopy (Scheme 10). Anal. Calcd for
C47H91N3Si3W: C, 58.42; H, 8.45; N, 4.35. Found: C, 58.52; H, 8.82;
N, 4.13.
1H, 13C{1H}, and 2H{1H} NMR spectra were obtained using Varian
XL-200, XL-400, and Unity-500 spectrometers. Chemical shifts are
reported relative to benzene-d6 (1H, δ 7.15; 13C, δ 128.00) or THF-d8
(1H, δ 3.58; 13C, δ 67.57). Infrared spectra were recorded on a Nicolet
Impact 410 spectrophotometer interfaced to a Gateway PC.
Combustion analyses were conducted by Oneida Research Services,
Whitesboro, NY, or Robertson Microlit Laboratories, Madison, NJ.
Procedures. 1. [(tBu3SiN)3WH]K (2-H-K) + CD3I (Excess).
(a). General Considerations. In a small glass bomb was loaded 2-H-
K, followed by evacuation and distillation of the appropriate solvent.
Roughly 10 equiv of CD3I was vacuum-transferred and the bomb
placed in a heated oil bath. At the completion of the reaction, the
volatiles were removed in vacuo and the resulting mixture was
analyzed by NMR spectroscopy. The ratio of 1-ND-CD2H to 1-CD3
was obtained by assuming the 2H{1H} resonance at δ 1.41
4. (tBu3SiNH)(tBu3SiN)2WCHCHnBu (1-CHCHnBu). A small
bomb reactor containing 2-H-K (0.920 g, 1.06 mmol) and 20 mL of
benzene was cooled to −78 °C, and 6-bromo-1-hexene (1.2 mL, 8.5
equiv) was admitted by vacuum transfer. The colorless solution was
heated at 100 °C for 24 h to give a yellow solution with a precipitate.
After removal of the volatiles the residue was subject to sublimation at
80 °C and 10−4 Torr to remove unreacted substrate. The remaining
6535
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
material was taken up in hexanes, filtered, and passed through a basic
alumina column (ICN, activity grade I, for removal of 2-Br-K). The
hexanes were removed and replaced by bis(trimethylsilyl) ether, and
the resulting solution was cooled to −45 °C, yielding 1-C(H)C(H)
nBu, which was collected as a yellow powder by filtration (240 mg,
25%). 1H NMR spectral analysis showed the Z to E ratio to be 7.1/1 in
addition to a small amount of 1-Br (<10%). IR (Nujol, cm−1): 3243
(w,NH), 1579 (w, CC), 1467 (s), 1386 (s), 1365 (m), 1138 (s),
1045 (s), 1009 (s), 934 (m), 848 (s), 819 (s), 621 (s), 581 (s). Anal.
Anal. Calcd for C40H87N3Si3W (a mixture of isomers): C, 54.70; H,
9.98; N, 4.78. Found: C, 54.51; H, 9.81; N, 4.61.
8. (tBu3SiNH)(tBu3SiN)2WSiMe3 (1-SiMe3). A small flask contain-
ing 2-H-K (0.500 g, 0.580 mmol) and 25 mL of benzene was cooled to
−78 °C, and bromotrimethylsilane (8.5 equiv) was admitted by
vacuum transfer. The colorless solution was stirred at 23 °C for 3 h,
forming a light yellow solution with undissolved precipitate. After
removal of the volatile components, the remaining material was taken
up in hexanes and filtered and hexanes replaced by bis(trimethylsilyl)
ether. The resulting solution was cooled to −78 °C to give a milky
suspension of 1-SiMe3, which was collected as a white powder by
filtration (70 mg, 13%). IR (Nujol, cm−1): 3222 (m), 1477 (s), 1470
(s), 1385 (s), 1364 (m), 1250 (w), 1238 (m), 1127 (s), 1036 (s), 1008
(s), 934 (m), 849 (s), 830 (s), 820 (s), 737 (m), 621 (s), 579 (s).
Anal. Calcd for C39H91N3Si4W: C, 52.14; H, 10.05; N, 4.68. Found: C,
52.25; H, 10.04; N, 4.69.
Calcd for C42H93N3Si W: C, 55.54; H, 10.32; N, 4.63. Found: C,
3
54.57; H, 9.90; N, 4.52.
5. (tBu3SiNH)(tBu3SiN)2WcPrMe (1-cPrMe). (a). Preparation. A
small bomb reactor containing 2-H-K (0.860 g, 0.995 mmol) and 25
mL of benzene was cooled to −78 °C, and cPrCH2Br (1.0 g, 7.4 equiv)
was admitted by vacuum transfer. The colorless solution was heated to
100 °C for 6 h, giving a light pink solution with a precipitate. After
removal of the volatiles the residue was subject to sublimation at 80 °C
at 10−4 Torr. The remaining material was taken up in hexanes, filtered,
and passed through a basic alumina column (ICN, activity grade I).
The hexanes were removed and replaced by bis(trimethylsilyl) ether,
and the resulting solution was cooled to −45 °C, yielding 1-cPrMe,
which was collected as an off-white powder by filtration (454 mg,
52%). A 1H−1H E.COSY experiment (Varian Unity 500 spectrometer;
500.518 MHz) was conducted, and 8192 complex points in F2 and
150 complex points in F1 were acquired. Zero filling in both
dimensions resulted in a final spectrum of 8192 × 512 real data points.
The digital resolution was 0.5 Hz/pt in the F2 dimension and 8.0 Hz/
pt in the F1 dimension. A total of 32 transients were collected for each
FID using a relaxation delay of 1.5 s. Phase-sensitive acquisition was
achieved by the hypercomplex method of States, Haberkorn, and
Ruben.50 IR (Nujol, cm−1): 3254 (w,NH), 1468 (s), 1385 (s), 1131
(m), 1044 (s), 1008 (s), 935 (m), 846 (s), 820 (s), 621 (s), 582 (m).
Anal. Calcd for C40H89N3Si3W: 54.58; H, 10.19; N, 4.77. Found: C,
NMR Tube Reactions. Flame-dried NMR tubes sealed to 14/20
ground glass joints were charged with 2-H-K (typically 10 mg, 1.2 ×
−5
10 mol) in the drybox and removed to the vacuum line on needle
valve adapters. The NMR tube was degassed, deuterated solvent was
vacuum-transferred, and the volatile reagents were introduced by
vacuum transfer via a calibrated gas bulb at −196 °C. The tubes were
then sealed with a torch. Solid substrates were loaded with 2-H-K
followed by solvent transfer on the vacuum line and flame-sealing. All
reactions were performed in C6D6 with an excess of alkyl halide,
usually ∼10 equiv. Consult the schemes and equations for product
distributions determined by 1H, 13C{1H}, and 2H{1H} NMR
spectroscopy. In certain circumstances, such as conversion of the 6/
1 3-MeCCMe to 4-MeCCMe mixture to 4-MeCCMe, the solids were
simply added to the NMR tube in the drybox, deuterated solvent was
added, and the tube was sealed under vacuum for thermolysis.
General Kinetics Procedures. Solutions of 2-H-M (M = Na, K)
were prepared with C6D6 in 2 mL volumetric flasks. Bis(trimethylsilyl)
ether (∼0.5 μL) was added as an internal integration standard. Three
samples of 0.6 mL each were transferred to flame-dried 5 mm NMR
tubes joined to 14/20 joints and attached to needle valves. The tubes
were subject to three freeze−pump−thaw degas cycles followed by
transfer of the CH3X (X = Cl, Br, I) substrate via a calibrated gas bulb
prior to flame-sealing under vacuum. For kinetic runs able to be
completed in 3 h or less, the NMR tubes were individually placed in
the NMR probe preheated to 60.0 °C and the reaction was allowed to
proceed. For longer runs, the three NMR tubes were simultaneously
submerged in a 60.0 °C polyethylene glycol bath with a Tamson
immersion circulator. In both cases the temperature was stable to ±0.3
°C. Rates of disappearance were monitored by measuring the
integrated intensity of the hydride tungsten satellite at δ 6.80 in the
case of both 2-H-K and 2-H-Na. Integration of the satellite was
required due to the obstruction of the central resonance by the
residual solvent protons. The low relative intensity of the satellites
necessitated using 16 transient spectra for data collection recorded at
measured time intervals. All runs were monitored for 3−4 half-lives
except the slowest (Table 2). Rates and uncertainties were obtained
from nonlinear, nonweighted least-squares fitting to the exponential
form of the rate expression, averaged for the three NMR tubes for each
run.
1
54.31; H, 10.35; N, 4.84. H NMR spectral analysis of the original
reaction mixture revealed the composition to be 91% 1-cPrMe and 9%
1-C(H)C(H)Et.
(b). NMR Tube Scale. 2-H-K and 10 equiv of cPrCH2Br were
1
heated at 100 °C for 6 h. H and 13C{1H} NMR spectral analysis
indicated formation of 86% 1-cPrMe, 4% 1-C(H)C(H)Et, 8% 2-Br-
K, 2% 1-Br, and 1-butene.
6. (tBu3SiNH)(tBu3SiN)2WCHCCH2 (1-CHCCH2). (a).
Preparation. Propargyl bromide (0.8 mL, 7.2 mmol, 12 equiv) was
condensed into a glass reaction vessel containing 2-H-K (520 mg,
0.602 mmol) and 20 mL of hexanes. The reaction mixture was stirred
at 60 °C for 8 h, and the colorless solution changed to light red.
Following removal of all volatiles, the remaining material was taken up
in hexanes, filtered, condensed to ∼3 mL, and cooled to −78 °C. The
resulting light yellow powder was collected by filtration to yield 160
mg of 1-CHCCH2 (31%). IR (Nujol, cm−1): 3249 (m, NH),
1915 (s, CCCasym), 1470 (s), 1387 (s), 1365 (s), 1143 (s), 1128
(s), 1045 (s), 1009 (s), 935 (s), 845 (s), 820 (s), 792 (m), 622 (s),
582 (s). Anal. Calcd for C39H85N3Si3W: C, 54.20; H, 9.91; N, 4.86.
Found: C, 53.87; H, 9.66; N, 4.91.
(b). NMR Tube Scale. A corresponding NMR tube scale reaction
with 0.008 g of 2-H-K (9.3 × 10−6 mol) and 10 equiv of propargyl
bromide was heated for 24 h at 60 °C and then assayed by NMR
spectroscopy (Scheme 12).
Single-Crystal X-ray Diffraction Studies. 1.
(tBu3SiN)2(tBu3SiNH)WCH3 (1-CH3). A colorless crystal (0.4 × 0.3
× 0.2 mm) of 1-CH3, obtained by slow cooling of a hexanes solution,
was mounted in a glass capillary and placed on a goniometer head on a
Siemens P4 four-circle diffractometer that utilized graphite-mono-
chromated Mo Kα radiation (λ = 0.710 73 Å). The data were collected
at 22 °C using an ω-scan with variable speed of 2−30°/min.
Preliminary diffraction data revealed a monoclinic crystal system, and a
hemisphere routine was used for data collection. The data consisted of
7. (tBu3SiN)2W(κ-N,C-N(SitBu3)C(Me)C(Me)−) (3-MeCCMe). 1-
Bromo-2-butyne (1.3 mL, 15 mmol, 15 equiv) was condensed into a
glass reaction vessel containing 2-H-K (850 mg, 0.995 mmol) and 20
mL of hexanes. The reaction mixture was stirred at 23 °C for 24 h. The
colorless solution changed to light red within 10 min and gradually
became deep red. Following removal of all volatiles, the remaining
material was taken up in hexanes, filtered, condensed to ∼3 mL, and
cooled to −78 °C for 10 h. The resulting purple powder was collected
by filtration to yield 320 mg (36%). Analysis by 1H NMR revealed a 6/
6726 reflections, 6027 of which were symmetry independent (Rint
=
3
1 ratio of 3-MeCCMe to (tBu3SiN)2W(κ-C,η -C,N-CH2CMe2Si-
0.0373) and 4651 were greater than 2σ(I). The structure was solved by
direct methods (SHELXS), completed by difference Fourier syntheses.
and refined by full-matrix least-squares procedures (SHELXL).51 The
(tBu2)NCMeCHMe) (4-MeCCMe). IR (Nujol, cm−1): 1478 (s, br),
1386 (s), 1363 (s), 1263 (w), 1185 (m), 1139 (s), 1116 (s), 1050 (s),
1009 (s), 987 (s), 934 (s), 845 (m), 818 (s), 715 (s), 618 (s), 581 (s).
refinement utilized w−1 = σ2(Fo ) + (0.0288p)2 + 2.0041p, where p =
2
6536
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
2
(Fo + 2Fc2)/3. All non-hydrogen atoms were refined anisotropically,
6336. (c) Young, R. D.; Hill, A. F.; Hillier, W.; Ball, G. E. J. Am. Chem.
Soc. 2011, 133, 13806−13809.
and hydrogen atoms were included at calculated positions.
2. (tBu3SiN)2(tBu3SiNH)W(cCHCH2CHMe) (1-cPrMe). A colorless
block (0.2 × 0.1 × 0.05 mm) of 1-cPrMe, obtained by slow
evaporation of an Et2O solution, was immersed in polyisobutylene,
extracted on a glass fiber, and placed under a 173 K nitrogen stream on
the goniometer head of a Siemens SMART CCD area detector system
equipped with a fine-focus Mo X-ray tube (λ = 0.710 73 Å).
Preliminary diffraction data revealed a triclinic crystal system, and a
hemisphere routine was used for data collection. The data were
processed via the Bruker SAINT program to afford 17 566 reflections,
10 170 of which were symmetry independent (Rint = 0.1330) and 3161
were greater than 2σ(I). The data was corrected for absorption with
SADABS, and the structure was solved by direct methods (SHELXS),
completed by difference Fourier syntheses, and refined by full-matrix
least-squares procedures (SHELXL).51 The data quality was poor and
the structure was initially solved in P1, until pairs of molecules were
found to be related by an inversion center; the structure was
(3) (a) Lawes, D. J.; Darwish, T. A.; Clark, T.; Harper, J. B.; Ball, G.
E. Angew. Chem., Int. Ed. 2006, 45, 4486−4490. (b) Ball, G. E.;
Brookes, C. M.; Cowan, A. J.; Darwish, T. A.; George, M. W.;
Kawanami, H. K.; Portius, P.; Rourke, J. P. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 6927−6932.
(4) (a) Bromberg, S. E.; Yang, H.; Asplund, M. C.; Lian, T.;
McNamara, B. K.; Kotz, K. T.; Yeston, J. S.; Wilkens, M.; Frei, H.;
Bergman, R. G.; Harris, C. B. Science 1997, 278, 260−263. (b) Lian,
T.; Bromberg, S. E.; Yang, H.; Prouix, G.; Bergman, R. G.; Harris, C. B.
J. Am. Chem. Soc. 1996, 118, 3769−3770. (c) Asplund, M. C.; Snee, P.
T.; Yeston, Y. S.; Wilkens, M. J.; Payne, C. K.; Yang, H.; Kotz, K. T.;
Frei, H.; Bergman, R. G.; Harris, C. B. J. Am. Chem. Soc. 2002, 124,
10605−10612.
(5) Sawyer, K. R.; Cahoon, J. F.; Shanoski, J. E.; Glascoe, E. A.; King,
M. F.; Schlegel, J. P.; Zoerb, M. C.; Hapke, M.; Hartwig, J. F.; Webster,
C. E.; Harris, C. B. J. Am. Chem. Soc. 2010, 132, 1848−1859.
(6) (a) Calladine, J. A.; Torres, O.; Ansley, M.; Ball, G. E.; Bergman,
R. G.; Curley, J.; Buckett, S. B.; George, M. W.; Gilson, A. I.; Lawes, D.
J.; Perutz, R. N.; Sun, X. Z.; Vollhardt, K. P. C Chem. Sci. 2010, 1,
622−630. (b) Calladine, J. A.; Duckett, S. B.; George, M. W.;
Matthews, S. L.; Perutz, R. N.; Torres, O.; Vuong, K. Q. J. Am. Chem.
Soc. 2011, 133, 2303−2310.
transformed to and refined in P1. The refinement utilized w−1
=
̅
σ2(Fo ) + (0.1151p)2 + 251.4963p, where p = (Fo + 2Fc2)/3. The
tungsten atoms were refined anisotropically, but the data quality
mandated an isotropic refinement for the remaining heavy atoms.
Hydrogens were included at calculated positions. Constraints were
2
2
t
applied to the Bu3Si fragments such that chemically equivalent inter-
and intraatomic distances utilized the same least-squares variables (i.e.,
all d(SiCq) values are equivalent; all d(CqC) values are equivalent,
etc.). A diffusely formed and positionally disordered small molecule
was located in an area remote from either compound near the center
of symmetry; it was treated and refined as four carbon atoms per
asymmetric unit.
(7) Churchill, D. G.; Janak, K. E.; Wittenberg, J. S.; Parkin, G. J. Am.
Chem. Soc. 2003, 125, 1403−1420.
(8) Gould, G. L.; Heinekey, D. M. J. Am. Chem. Soc. 1989, 111,
5502−5504.
(9) Gross, C. L.; Girolami, G. S. J. Am. Chem. Soc. 1998, 120, 6605−
6606.
3. (tBu3SiN)2(tBu3SiNH)W(CHCCH2) (1-CHCCH2). A colorless
block (0.2 × 0.15 × 0.1 mm) of 1-CHCCH2, obtained by slow
evaporation of a hexane solution, was immersed in polyisobutylene,
extracted with a glass fiber, and placed under a 173 K nitrogen stream
on the goniometer head of a Siemens SMART CCD area detector
system equipped with a fine-focus Mo X-ray tube (λ = 0.710 73 Å).
Preliminary diffraction data revealed a monoclinic crystal system, and a
hemisphere routine was used for data collection. The data were
processed via the Bruker SAINT program to afford 27 007 reflections,
10 355 of which were symmetry independent (Rint = 0.0663) and 5347
were greater than 2σ(I). The data were corrected for absorption with
SADABS, and the structure was solved by direct methods (SHELXS),
completed by difference Fourier syntheses, and refined by full-matrix
(10) Bernskoetter, W. H.; Hanson, S. K.; Buzak, S. K.; Davis, Z.;
White, P. S.; Swartz, R.; Goldberg, K. I.; Brookhart, M. J. Am. Chem.
Soc. 2009, 131, 8603−8613.
(11) Jones, W. D. Acc. Chem. Res. 2003, 36, 140−146.
(12) Evans, D. R.; Drovetskaya, T.; Bau, R.; Reed, C. A.; Boyd, P. D.
W. J. Am. Chem. Soc. 1997, 119, 3633−3634.
(13) Castro-Rodriguez, I.; Nakai, H.; Gantzel, P.; Zaknarov, L. N.;
Rheingold, A. L.; Meyer, K. J. Am. Chem. Soc. 2003, 125, 15734−
15735.
(14) Wolczanski, P. T. Chem. Commun. 2009, 740−757.
(15) Cundari, T. R.; Klinckman, T. R.; Wolczanski, P. T. J. Am. Chem.
Soc. 2002, 124, 1481−1487.
(16) (a) Cundari, T. R. J. Am. Chem. Soc. 1992, 114, 10557−10563.
(b) Cundari, T. R. Organometallics 1993, 12, 4971−4978. (c) Cundari,
T. R. Organometallics 1994, 13, 2987−2994.
least-squares procedures (SHELXL).51 The refinement utilized w−1
=
σ2(Fo ) + (0.0149p)2 + 3.287p, where p = (Fo + 2Fc2)/3. All non-
hydrogen atoms were refined anisotropically, and hydrogen atoms
were included at calculated positions.
2
2
(17) (a) Benson, M. T.; Cundari, T. R.; Moody, E. W. J. Organomet.
Chem. 1995, 504, 1−13. (b) Cundari, T. R.; Klinckman, T. R. Inorg.
Chem. 1998, 37, 5399−5401. (c) Cundari, T. R.; Matsunaga, N.;
Moody, E. W. J. Phys. Chem. 1996, 100, 6475−6483.
(18) Cundari, T. R.; Pierpont, A. W.; Rabaa, H. Int. J. Quantum Chem.
2006, 106, 1611−1619.
ASSOCIATED CONTENT
■
S
* Supporting Information
CIF files giving crystal data for 1-CH3, 1-cPrMe, and 1-CH
CCH2. This material is available free of charge via the
Internet at http://pubs.acs.org.
(19) Bennett, J. L.; Wolczanski, P. T. J. Am. Chem. Soc. 1994, 116,
2179−2180.
(20) (a) Bennett, J. L.; Wolczanski, P. T. J. Am. Chem. Soc. 1997, 119,
10696−10719. (b) Bennett, J. L.; Vaid, T. P.; Wolczanski, P. T. Inorg.
Chim. Acta 1998, 270, 414−423.
AUTHOR INFORMATION
Corresponding Author
*Fax: 607 255 4173. E-mail: ptw2@cornell.edu.
■
(21) Slaughter, L. M.; Wolczanski, P. T.; Klinckman, T. R.; Cundari,
T. R. J. Am. Chem. Soc. 2000, 122, 7953−7975.
(22) Cummins, C. C.; Schaller, C. P.; Van Duyne, G. D.; Wolczanski,
P. T.; Chan, E. A.-W.; Hoffmann, R. J. Am. Chem. Soc. 1991, 113,
2985−2994.
(23) (a) Schaller, C. P.; Cummins, C. C.; Wolczanski, P. T. J. Am.
Chem. Soc. 1996, 118, 591−611. (b) Cummins, C. C.; Baxter, S. M.;
Wolczanski, P. T. J. Am. Chem. Soc. 1988, 110, 8731−8733.
(24) Schaller, C. P.; Bonanno, J. B.; Wolczanski, P. T. J. Am. Chem.
Soc. 1994, 116, 4133−4134.
(25) de With, J.; Horton, A. D. Angew. Chem., Int. Ed. Engl. 1993, 32,
903−905.
ACKNOWLEDGMENTS
P.T.W. thanks the NSF (Nos. CHE-9528914 and CHE-
1055505) and Cornell University for financial support.
■
REFERENCES
■
(1) Bernskoetter, W. H.; Schauer, C. K.; Goldberg, K. I.; Brookhart,
M. Science 2009, 326, 553−556.
(2) (a) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1998, 120, 9953−
9954. (b) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1999, 121, 6336−
6537
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538

Organometallics
Article
(26) de With, J.; Horton, A. D.; Orpen, A. G. Organometallics 1993,
12, 1493−1496.
(27) Schaller, C. P.; Wolczanski, P. T. Inorg. Chem. 1993, 32, 131−
144.
(28) Schafer, D. F. II; Wolczanski, P. T. J. Am. Chem. Soc. 1998, 120,
4881−4882.
(29) Schafer, D. F., II Ph.D. Thesis, Cornell University, 1999.
(30) Schafer, D. F., II; Wolczanski, P. T.; Lobkovsky, E. B.
Organometallics 2011, 30, DOI: 10.0121/om200597z.
(31) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in
Organic Chemistry, 3rd ed.; Harper and Row: New York, 1987.
(32) Wigley, D. E. Prog. Inorg. Chem. 1994, 42, 239−482.
(33) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; Wiley:
New York, 1988.
(34) Carpenter, B. K. Determination of Reaction Mechanisms; Wiley-
Interscience: New York, 1984.
(35) (a) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317−323.
(b) Newcomb, M.; Toy, P. H. Acc. Chem. Res. 2000, 33, 449−455.
(36) Bowry, V. W.; Lusztyk, J.; Ingold, K. U. J. Am. Chem. Soc. 1991,
113, 5687−5698.
(37) (a) Chatgilialoglu, C.; Ingold, K. U.; Scaiano, J. C. J. Am. Chem.
Soc. 1981, 103, 7739−7742. (b) Schmid, P.; Griller, D.; Ingold, K. U.
Int. J. Chem. Kinet. 1979, 11, 333−338.
(38) (a) Hoyt, H. M.; Bergman, R. G. Angew. Chem., Int. Ed. 2007,
46, 5580−5582. (b) Hoyt, H. M.; Michael, F. E.; Bergman, R. G. J. Am.
Chem. Soc. 2004, 126, 1018−1019. (c) Lee, S. Y.; Bergman, R. G. J.
Am. Chem. Soc. 1995, 117, 5877−5878. (d) Walsh, P. J.; Hollander, F.
J.; Bergman, R. G. J. Am. Chem. Soc. 1988, 110, 8729−8731.
(39) Jones, W. D. Inorg. Chem. 2005, 44, 4475−4484.
(40) Ng, S. H. K.; Adams, C. S.; Hayton, T. W.; Legzdins, P.; Patrick,
B. O. J. Am. Chem. Soc. 2003, 125, 15210−15223.
(41) (a) Tseng, T. W.; Wu, I. Y.; Tsai, J. H.; Lin, Y. C.; Chen, D. J.;
Lee, G. H.; Cheng, M. C.; Wang, Y. Organometallics 1994, 13, 3963−
3971. (b) Keng, R. S.; Lin, Y. C. Organometallics 1990, 9, 289−291.
(42) (a) Ang, W. H.; Cordiner, R. L.; Hill, A. F.; Perry, T. L.; Wagler,
J. Organometallics 2009, 28, 5568−5574. (b) Bussetto, L.; Marchetti,
F.; Zacchini, S.; Zanotti, V. Eur. J. Inorg. Chem. 2005, 3250−3260.
(c) Wells, M. B.; McConathy, J. E.; White, P. S.; Templeton, J. L.
Organometallics 2002, 21, 5007−5020. (d) Chetcuti, M. J.; McDonald,
S. R. Organometallics 2002, 21, 3162−3168. (e) Ogoshi, S.; Nishida,
T.; Fukunishi, Y.; Tsutsumi, K.; Kurosawa, H. J. Organomet. Chem.
2001, 620, 190−193.
(43) Polse, J. L.; Andersen, R. A.; Bergman, R. G. J. Am. Chem. Soc.
1998, 120, 13405−13414.
(44) Rosenfeld, D. C.; Wolczanski, P. T.; Barakat, K. A.; Buda, C.;
Cundari, T. R.; Schroeder, F. C.; Lobkovsky, E. B. Inorg. Chem. 2007,
46, 9715−9735.
(45) (a) Mayer, J. M. J. Phys. Chem. Lett. 2011, 2, 1481−1489.
(b) Warren, J. T.; Mayer, J. M. Proc. Natl. Acad. Sci. U.S.A. 2010, 107,
5282−5287.
(46) Manner, V. W.; Lindsay, A. D.; Mader, E. A.; Harvey, J. N.;
Mayer, J. M. Chem. Sci. 2011, in press.
(47) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mill Valley, CA, 1987.
(48) (a) Williams, G. M.; Gell, K. I.; Schwartz, J. J. Am. Chem. Soc.
1980, 102, 3660−3662. (b) Gell, K. I.; Schwartz, J. J. Am. Chem. Soc.
1981, 103, 2687−2695.
(49) (a) Adams, C. S.; Legzdins, P.; Tran, E. J. Am. Chem. Soc. 2001,
123, 612−624. (b) Adams, C. S.; Legzdins, P.; McNeil, W. S.
Organometallics 2001, 20, 4939−4955. (c) Adams, C. S.; Legzdins, P.;
Tran, E. Organometallics 2002, 21, 1474−1486.
(50) States, D. J.; Haberkorn, R. A.; Ruben, D. J. J. Magn. Reson.
1982, 48, 286−289.
(51) Sheldrick, G. M. SHELXTL: An Integrated System for Solving,
Refining and Displaying Crystal Structures from Diffraction Data;
University of Gottingen, Gottingen, Germany, 1981.
6538
dx.doi.org/10.1021/om2009342 | Organometallics 2011, 30, 6518−6538