Detection of unusual reaction intermediates during the conversion of W(N2)2(dppe)2 to W(H)4(dppe) 2 and of H2O into H2

Article abstract of DOI:10.1021/ja302202q

W(N₂)₂(dppe-κ²P)₂ reacts with H₂ to form WH₃{Ph(C₆H₄)PCH ₂CH₂PPh₂-κ²P}(dppe- κ²P) and then W(H)₄(dppe-κ²P) ₂. When para-hydrogen is used in this study, polarized hydride signals are seen for these two species. The reaction is complicated by the fact that trace amounts of water lead to the formation of H₂, PPh ₂CH₂CH₂Ph₂P(O) and W(H) ₃(OH)(dppe-κ²P)₂, the latter of which reacts further via H₂O elimination to form W(H)₄(dppe- κ²P)₂ and [WH₃{Ph(C₆H ₄)PCH₂CH₂PPh₂-κ²P} (dppe-κ²P)]. These studies demonstrate a role for the 14-electron intermediate W(dppe-κ²P)₂ in the CH activation reaction pathway leading to [WH₃{Ph(C₆H ₄)PCH₂CH₂PPh₂-k²P}(dppe- k²P)]. UV irradiation of W(H)₄(dppe-κ²P) ₂ under H₂ led to phosphine dechelation and the formation of W(H)₆(dppe-k²P)(dppe-k¹P) rather than H ₂ loss and W(H)₂(dppe-κ²P)₂ as expected. Parallel DFT studies using the simplified model system W(N ₂)₂((Ph)HPCH₂CH₂PH ₂-κ²P)(H₂PCH₂CH ₂PH₂-κ²P) confirm that ortho-metalation is viable via both W(dppe-κ²P)₂ and W(H) ₂(dppe-κ²P)₂ with explicit THF solvation being necessary to produce the electronic singlet-based reaction pathway that matches with the observation of para-hydrogen induced polarization in the hydride signals of [WH₃{Ph(C₆H₄)PCH ₂CH₂PPh₂-κ²P}(dppe- κ²P)], W(H)₃(OH)(dppe-κ²P) ₂ and W(H)₄(dppe-κ²P)₂ during this study. These studies therefore reveal the existence of differentiated and previously unsuspected thermal and photochemical reaction pathways in the chemistry of both W(N₂)₂(dppe-κ²P) ₂ and W(H)₄(dppe-κ²P)₂ which have implications for their reported role in N₂ fixation.

Full text of DOI:10.1021/ja302202q

Article
pubs.acs.org/JACS
Detection of Unusual Reaction Intermediates during the Conversion
of W(N2)2(dppe)2 to W(H)4(dppe)2 and of H2O into H2
Beatriz Eguillor,† Patrick J. Caldwell,† Martin C. R. Cockett,† Simon B. Duckett,*,† Richard O. John,†
Jason M. Lynam,† Christopher J. Sleigh,† and Ian Wilson‡
†Department of Chemistry, University of York, Heslington, York , U.K. YO10 5DD
‡Department of Safety of Medicines, AstraZeneca, Alderley Park, Macclesfield, Cheshire, U.K. SK10
S
* Supporting Information
ABSTRACT: W(N2)2(dppe-κ2P)2 reacts with H2 to form WH3{Ph-
(C6H4)PCH2CH2PPh2-κ2P}(dppe-κ2P) and then W(H)4(dppe-κ2P)2.
When para-hydrogen is used in this study, polarized hydride signals
are seen for these two species. The reaction is complicated by the fact
that trace amounts of water lead to the formation of H2,
PPh2CH2CH2Ph2P(O) and W(H)3(OH)(dppe-κ2P)2, the latter of
which reacts further via H2O elimination to form W(H)4(dppe-κ2P)2
and [WH3{Ph(C6H4)PCH2CH2PPh2-κ2P}(dppe-κ2P)]. These stud-
ies demonstrate a role for the 14-electron intermediate W(dppe-κ2P)2
in the CH activation reaction pathway leading to [WH3{Ph(C6H4)-
PCH2CH2PPh2-k2P}(dppe-k2P)]. UV irradiation of W(H)4(dppe-
κ2P)2 under H2 led to phosphine dechelation and the formation of
W(H)6(dppe-k2P)(dppe-k1P) rather than H2 loss and W(H)2(dppe-
κ2P)2 as expected. Parallel DFT studies using the simplified model system W(N2)2((Ph)HPCH2CH2PH2-κ2P)-
(H2PCH2CH2PH2-κ2P) confirm that ortho-metalation is viable via both W(dppe-κ2P)2 and W(H)2(dppe-κ2P)2 with explicit
THF solvation being necessary to produce the electronic singlet-based reaction pathway that matches with the observation of
para-hydrogen induced polarization in the hydride signals of [WH3{Ph(C6H4)PCH2CH2PPh2-κ2P}(dppe-κ2P)], W(H)3(OH)-
(dppe-κ2P)2 and W(H)4(dppe-κ2P)2 during this study. These studies therefore reveal the existence of differentiated and
previously unsuspected thermal and photochemical reaction pathways in the chemistry of both W(N2)2(dppe-κ2P)2 and
W(H)4(dppe-κ2P)2 which have implications for their reported role in N2 fixation.
ment of a second metal has also been shown to facilitate proton
transfer.14,15
INTRODUCTION
■
Systems such as W(N2)2(dppe-κ2P)2 (1) and W(CO)3(PCy3)2
provide important examples of complexes where the activation
of small molecules by inorganic compounds can be seen in
action.1 In the case of W(N2)2(dppe-κ2P)2, the binding of
dinitrogen and the resulting sensitivity to electrophilic attack
have particular relevance to nitrogen fixation.2 W-
(CO)3(PCy3)2, on the other hand, has provided a model
system to enable the subsequent rationalization of dihydrogen
activation in terms of both classical dihydride and dihydrogen
interactions that are now ubiquitous to inorganic chemistry.3−6
The chemistry of tungsten polyhydride complexes containing
mono and bidentate phosphine ligands has also been the
subject of significant investigation,16 with many dihydride,
tetrahydride and hexahydride complexes reported. However, as
in the case of related polyhydride iridium systems,17 the mixing
of η2-H2 and M(H)2 coordination modes adds significantly to
the complexity of ligand binding. This results in an array of
W(0)−W(VI) complexes and difficulties in their unambiguous
characterization. Utilization of T1(min) values as a route to
achieve this differentiation is well established.18,19 For example,
NMR data for WH6(PPh(CH2CH2PPh2)2 has produced T1
values of 103 and 126 ms at 223 K which suggests classical
hexahydride formulation.20 The protonated form of this
complex, [W(H)7−2x(η2-H2)x(PPh(CH2CH2PPh2)2]BF4 (x =
1 or 2) was, however, found to yield a T1 value of 21 ms at 223
K which is consistent with the involvement of exchanging η2-H2
ligands. Related protonation studies on the classical tetrahy-
dride W(H)4(PMePh2)4 lead to initial attack on the hydrido
7
Reactivity studies involving W(N2)2(dppe-κ2P)2 suggest the
generation of coordinatively unsaturated W(dppe-κ2P)2 which
has led to a range of novel reactions. These include the removal
of CO from dimethylformide,8 and the conversion of
benzaldehydeiminines into isocyanides.9 Studies have also
demonstrated that the dppe ligand can be converted by this
complex into the tetradentate phosphine meso-o-
C6H4(PPhCH2CH2PPh2)2.10 Furthermore, unusual reactivity
has also been observed in other tungsten phosphine systems.11
Others have demonstrated that protonation of W(N2)2(dppe-
κ2P)2 leads to a hydrazido complex, with ammonia
subsequently formed under mild conditions.12,13 The involve-
Received: March 6, 2012
Published: October 16, 2012
© 2012 American Chemical Society
18257
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
1H NMR (700 MHz, THF-d8, 283 K): δ 7.90 (bs, JHH = 7.3 Hz, 8H,
Ph-ortho), 7.21 (t, JHH = 7.3 Hz, 8H, Ph-meta), 7.17 (m, JHH = 7.0 Hz,
8H, Ph-ortho), 7.15 (m, JHH = 7.3 Hz, 4H, Ph-para), 6.76 (m, JHH = 7.0
ligand to generate [WH3(η2-H2)(PMePh2)4]+ which subse-
quently rearranges to give the pentahydride.21,15
Factors that control the binding of small molecules to
W(CO)3(PCy3)2 and its agostic bonding within the cyclohexyl
moiety have been explored by density functional theory
(DFT).22 The use of DFT to probe aromatic C−H bond
activation at tungsten23 and to examine polyhydride systems
has also been shown to enable rationalization of the reactivity
of such systems.24 Others have modeled the complex
functionalization of N2 by H2 in such organometallic systems.25
Furthermore, the interaction of hydride ligands with water has
been shown to result in the generation of a metal−hydroxide
complex and dihydrogen.26 Such DFT methods are particularly
useful in predicting the reactivity of reaction intermediates that
are normally experimentally undetectable.
NMR spectroscopy, in conjunction with UV photolysis of
samples within the NMR probe, has enabled the character-
ization of very unstable materials and the monitoring of their
reactivity. This approach has been used successfully to detect
unstable alkyl hydride27,28 complexes, a xenon complex29 and a
number of hydride complexes.30 More recently, the approach
has been used to generate unstable tungsten dihydrogen
complexes.31 The related technique of photo-Chemically
Induced Dynamic Nuclear Polarization (photo-CIDNP) has
been used to determine kinetic parameters for protein
folding.32
The in situ photochemical method has been extended to
enable the successful detection of materials that are unstable
even at 213 K through the signal enhancement resulting from
the para-hydrogen (para-H2)33,34 induced polarization effect
(PHIP),33−35 an approach which has been reviewed recently.36
In the present paper, we report first on a study of
W(H)2(CO)3(PCy3)2 using the in situ photolysis method
whose objective was to establish whether equilibration between
a metal η2-H2 complex and its dihydride counterpart could be
followed using PHIP. The reactivity of the complex W-
(N2)2(dppe-κ2P)2 (1) toward H2 is then followed using PHIP
and a number of interesting observations are described. The
resulting complexes are characterized by full NMR analysis
including 183W NMR data.37,38 These studies are rationalized
through model DFT calculations based on W(N2)2((Ph)-
HPCH2CH2PH2-κ2P)(H2PCH2CH2PH2-κ2P) (1m) where the
dppe ligand is replaced by a simplified phosphine.
Hz, 8H, Ph-meta), 6.79 (t, JHH = 7.0 Hz, 4H, Ph-para), 2.68 (dd, JHP
=
28.6 Hz, JHH = 9 Hz, 4H, −CH2−), 1.70 (m, JHH = 9 Hz, 4H,
−CH2−), −3.68 (binomial quintet, JHP = 26 Hz, JWH = 38 Hz, 4H).
13C{1H} NMR (176.008 MHz, THF-d8, 283K): δ 143.81 (d, JCP
=
39.7 Hz Cipso,), 139.26 (d, JCP = 32.5 Hz Cipso), 137.19 (Ph-ortho),
134.44 (Ph-ortho), 128.8 (Ph-para), 127.35 (Ph-meta), 127.14 (Ph-
para), 127.03 (Ph-meta) 34.76 (d, JCP = 32.5 Hz −CH2−). 31P{1H}
NMR (283.4 MHz, THF-d8, 283 K): 61.5 s (JWP = 169 Hz). T1(min)
(WH, 400 MHz, C7D8): 79 1.1 ms (signal at δ −3.63, temperature
273 K). 183W{1H} NMR (16.6 MHz, C7D8, 298 K): δ −3354.5 (m,
JWP = 167.3 Hz).
NMR Scale Preparation of [W(H)3{Ph(C6H4)PCH2CH2PPh2-
κ2P}(dppe-κ2P)] (3). A sample of 1 was dissolved in toluene-d8 and
placed in a dry ice/isopropanol bath at ca. 230 K. The tube was then
pressurized to 3 atm with hydrogen and the resulting sample irradiated
at 223 K using a broadband 180 W Oriel UV lamp for 40 min. At this
point NMR spectroscopy revealed that the sample contained 3 (see
below) at >90% conversion. This process enabled the characterization
of 3 to be completed at low temperature by COSY, NOE and HMQC
methods where its lifetime was sufficiently extended. Full NMR data
for 3 are presented in Tables 1 and s1. 183W{1H} NMR (16.6 MHz,
C7D8, 298 K): δ −2917.
Table 1. Partial NMR Data for 3 in Toluene-d8, 298 K
coupling
to Pa
(J/Hz)
coupling
to Pb
(J/Hz)
coupling
to Pc
(J/Hz)
coupling
to Pd
(J/Hz)
group, chemical shift /δ,
J (/Hz) (T1)
Ha, −0.14, JHaHc = 2, JHaHb
= 4 (T1 773 ms at 248
K)
24
68
34
8
Hb, −2.69, JHbHc = 8 (T1
62
6
26
24
12
64
22
38
724 ms at 248 K)
Hc, −2.93, JHH as above
(T1 510 ms at 248 K)
Pb, 59.3, JWP = 138
Pc, 47.4, JWP = 170
Pa, 55.4, JWP = 180
Pd, −13.2, JWP = 122
8
76
-
--
15
8
15
--
8
8
76
8
22
-
22
8
EXPERIMENTAL SECTION
■
General Conditions and Materials. All manipulations were
carried out under inert atmosphere conditions, using standard Schlenk
techniques (with N2 or Ar as the inert atmosphere) or high vacuum
techniques. Solvents were obtained as analytical grade from Fisher.
The WCl6 used was obtained from Acros Organics and the bis-(1,2-
diphenyl-phosphino)ethane (dppe) was obtained from Sigma-Aldrich
and used without further purification. W(N2)2(dppe-κ2P)2 (1) was
prepared by known literature methods7 and characterized by NMR.
NMR spectra were collected on Bruker DRX 400, Avance 600 and
Avance II 700 spectrometers.
NMR Scale Preparation of W(H)6(dppe-κ2P)(dppe-κ1P) (4). A
sample of 2 in toluene-d8 was cooled to 260 K, and then pressurized to
3 atm by hydrogen gas. The cooled sample was then exposed to UV
irradiation (Oriel lamp) for 120 min. At this point, NMR spectroscopy
revealed the sample to contain both 4 (see below) and 2 with no
evidence for 3 even though it had already been demonstrated to be
sufficiently thermally and photochemically stable under these
conditions. The ratio of 4 to 2 was found to be 1:1 at this point
and remained unchanged at this level after 30 min of further
irradiation. This route enabled the subsequent characterization of 4 at
Preparation and Characterization of W(H)4(dppe-κ2P)2 (2).
An NMR sample of 1 was prepared by charging a 5 mm NMR tube
fitted with a Young’s tap with ca. 1 mg of 1 and then adding toluene-d8
via vacuum transfer. Hydrogen (3 atm) was then added to this sample
at room temperature. After 2 h of heating at 328 K, the 31P{1H} NMR
spectrum of the resultant solutions showed only a singlet at 61.0 ppm
due to the known complex 2. The room temperature NMR data for 2
agree with that in the literature.39 We note an increase in the T1 value
to 585 ms when recorded with 31P decoupling at 258 K, in accordance
with a classical tetrahydride configuration for 2.
1
273 K by COSY, NOE and HMQC methods. H NMR (700 MHz,
toluene-d8, 273 K): δ 2.24 (m, JPH = 18 Hz, Hb) δ 7.92 (ortho), δ 7.04
18258
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
(meta), δ 7.00 (para), δ 2.67 (Hc) δ 7.82 (ortho), δ 6.98 (meta), δ
7.00 (para), δ 2.79 (m, Hd) δ 7.38 (ortho), δ 7.00 (meta), δ 7.03
(para); δ −2.41 (dq, JWH = 38.1 Hz, JHP(a) = 36.3 Hz, JHP(b) = 26.5 Hz,
Ha; 13C{1H} NMR (176.008 MHz, toluene-d8, 273K), attached to Pa,
δ 31.85 (d) JCP = 47 Hz, δ 139.52 (ipso), δ 133.12 (ortho), δ 128.04
(meta), δ 129.04 (para), attached to Pb, δ 34.60 (d) JCP = 18 Hz, δ
139.41(ipso), δ 132.96 (ortho), δ 127.23 (meta), δ 128.27 (para),
attached to Pc, δ 34.80 (d) JCP = 33 Hz, δ 132.93 (ortho), δ 129.09
(meta), δ 128.02 (para). 31P{1H} NMR (283.4 MHz, toluene-d8,
273K), PA, δ 54.7, JWH = 95 Hz, JP(a)P(b) = 7.4 Hz, PB, δ 42.3, JWH = 62
Hz, JP(a)P(b) = 7.4 Hz, JP(b)P(c) = 38 Hz, PC, δ −12.4, JP(b)P(c) = 38 Hz;
(16.6 MHz, C7D8, 298 K), T1(min) (WH, C7D8): 890 ms (signal at δ
−2.41, temperature 228 K). 183W{1H} NMR (16.6 MHz, C7D8, 298
K), δ −4000 (m).
the rapid relaxation associated with the η2-H2 isomer quenches
the PHIP effect in this system during tautomerization between
the W(H2)(CO)3(PCy3)2 and W(H)2(CO)3(PCy3)2 forms.
This result contrasts with that previously reported Bargon et al.
who showed that an organic hydrogenation product can show
PHIP even though it is formed via a metal-based intermediate
that exists in an η2-H2 form.40 In both cases, the lifetime of the
M-η2-H2 intermediate must play a pivotal role in determining
the extent of polarization in any newly formed hydrogenation
product. Consequently, we decided to extend the study to
target W(H)4(dppe-κ2P)2 (2).
NMR Scale Preparation of W(H)3(OH)(dppe-κ2P)2 (5). A
sample of 1 was dissolved in undried toluene-d8 and left to stand
overnight at room temperature. Three atm. of hydrogen was then
added and left to react for a further 24 h. During this period, NMR
signals due to 5 (see below) became evident. The ratio of 5 to 1
typically reached 1:5 after 24 h. Over a longer time period, the slow
thermal conversion of 5 into both 4 and 2 (ratio ca. 1:20) occurred.
Signals for compound 3 were also observed, in addition to those of
free PPh2CH2CH2PPh2(O). This method allowed the production of
sufficient 5 for its full NMR characterization at 283 K by COSY, NOE
and HMQC methods. A sample where 5 is the dominant product can
be obtained from a similar reaction in THF when it is doped with
H2O. Full NMR data for 5 are presented in Tables 2 and s2
Reactions of W(N2)2(dppe-κ2P)2 (1) with para-Hydro-
gen. W(N2)2(dppe-κ2P)2 (1) reacts with H2 to form
W(H)4(dppe-κ2P)2 (2),41 and when this reaction was followed
1
in THF-d8 with para-H2 at 333 K, the resulting H NMR
spectrum contains a polarized hydride signal at δ −3.66 as
shown in Figure 1a. This hydride signal appears as a quintet
Table 2. Partial NMR Data for 5
Figure 1. Selected 1H NMR signals observed during reactions of
W(N2)2(dppe-κ2P)2 (1) and H2. (a) PHIP polarized hydride signal for
W(H)4(dppe-κ2P)2 (2), (b) hydride signals, Ha, Hb and Hc of
[WH3{Ph(C6H4)PCH2CH2PPh2-κ2P}(dppe-κ2P)] (3).
group, chemical, shift/δ
(multiplicity) and T1 at
268 K
coupling
to Pa
(J/Hz)
coupling
to Pb
coupling
to Pc
coupling
to Pd
(J/Hz)
(J/Hz)
(J/Hz)
Ha, −2.26 (t); 284 ms
Hb, −2.37 (dd); 282 ms
Hc, 1.59 (ddd)
18
62
30
38
44
62
18
30
30
24
86
114
with JPH = 28 Hz and is flanked by 183W satellites with JHW = 38
Hz. A 2D 1H−183W HMQC37 spectrum was then recorded and
showed a 31P coupled multiplet at δ −3363 when referenced to
TMS at 100 MHz.42
OH, −3.00 (bs); 744 ms
Pb, 64.0 (bs)
16
-
16
-
21.6
14
Pc, 45.7 (AB spin system
JAB 152, Δν 496.5 Hz)
152
14
The origin of the PHIP enhancement seen in the NMR
signal observed for the hydride ligands of 2 results from the
second order [AX]4 spin system which provides the magnetic
inequivalence necessary to achieve PHIP activity.43 Interest-
ingly, weak PHIP polarization is also seen in one of the two sets
of ortho-phenyl proton resonances of 2 that appears at δ 7.42
where JPH = 8 Hz; the antiphase coupling reflected in the PHIP
enhancement is −3 Hz. This observation provides insight into
the intramolecular activation of the dppe-ligand (q.v.). The
corresponding 1H COSY NMR spectrum of 2 contained a weak
connection between the hydride resonance and the dppe
backbone H signals at δ 2.64 in addition to the ortho-phenyl
proton signal at δ 7.42 thereby confirming that they are all
weakly scalar coupled.
Previous studies have shown that PHIP can be used to
examine successfully a range of mono-, di- and trihydride
Pa, 47.6 (AB spin system)
-
14
152
9
14
-
Pd, 36.9 (bt)
9
21.6
RESULTS AND DISCUSSION
Photochemical Reactions of W(CO)3(PCy3)2 with para-
■
Hydrogen. A sample of W(CO)3(PCy3)2 was prepared and
then reacted with para-H2, with the ensuing thermal reaction
monitored by 1H NMR spectroscopy. While hydride signals for
the dihydride isomer, W(H)2(CO)3(PCy3)2, were observed at
298 K, they failed to exhibit any PHIP. We then employed the
in situ UV photolysis method to monitor this reaction directly
at 213 K using 325 nm radiation. However, while the hydride
signals for W(H)2(CO)3(PCy3)2 were detected, no PHIP was
seen in the two hydride resonances. We conclude therefore that
1
18259
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
complexes.44 The hydride ligand polarization observed in 2 and
reported here represents the first time that PHIP has been seen
in such a mononuclear tetrahydride complex. The T1(min) value
of the hydride signal of 2 was determined as 0.58 s at 258 K and
is fully consistent with the observation of PHIP in this species.
We conclude therefore that any W-η2-H2 based reaction
intermediate which might exist on the reaction pathway to 2
has a very short lifetime (see DFT discussion later).20,45
We expected this reaction to proceed via the intermediate
W(H)2(N2)(dppe-κ2P)2 but this was not detected at 333 K. We
therefore re-examined the reaction at the lower temperature of
295 K where we observed two further sets of polarized hydride
signals in the corresponding 1H NMR spectrum at δ −2.69 and
δ −2.93 (see Figure 1b). These resonances are due to the new
complex, 3, and proved to couple to a third hydride resonance
at δ −0.14 according to COSY methods. These data therefore
preclude the detection of W(H)2(N2)(dppe-κ2P)2 and suggest
that an unexpected reaction has taken place.
Figure 2. PHIP enhanced NMR signals observed during the
generation of W(H)4(dppe-κ2P)2 (2) and [WH3{Ph(C6H4)-
PCH2CH2PPh2-κ2P}(dppe-κ2P)] (3) under in situ photolysis at 298 K.
possess 1:1:1 integrals with 8 further distinct ethane bridge
protons also evident. Examination of a series of NMR spectra
revealed 39 aromatic proton resonances for 3 which arise from
seven sets of five coupled signals due to a phenyl ring and one
set of four coupled signals due to a CH bond activated phenyl
ring. In addition to this, 3 yields four 31P NMR signals that are
located at δ −13.2, 47.4, 55.4 and 59.3.
Photochemical Reactions of W(N2)2(dppe-κ2P)2 (1)
with para-Hydrogen. We describe now how the identity of
3, an intramolecular CH bond activation product, was
confirmed through its formation in a series of low temperature
irradiation studies. The formation of 2 and 3 from 1 is
illustrated in Scheme 1. The 297 nm irradiation of W-
The 31P NMR signal for free dppe appears at δ −12.8, close
to one of the resonances of 3 at δ −13.2. This suggests that 3
may contain a phosphorus center that is no longer coordinated
to the metal. However, the resonance at δ −13.2 couples
strongly (64 Hz) to the hydride ligand that resonates at δ
Scheme 1. Reaction Products and Pathways Involved in the
Reactions of W(N2)2(dppe-κ2P)2 (1) with H2
1
−2.63 in the H NMR spectrum of 3. It is therefore clear that
the phosphorus center giving rise to this signal is bound to the
metal. Such a high-field signal for a metal-bound phosphorus
atom has precedence as similar shifts are observed for metallo-
phosphorus 4-membered rings.47
A
13C resonance at δ 122.33
in 3 corresponds to a quaternary carbon directly attached to the
tungsten center and supports this ortho-metalation concept.
We conclude therefore that 3 is the trihydride complex
WH3{Ph(C6H4)PCH2CH2PPh2-κ2P}(dppe-κ2P) where one of
the phosphine ligands is ortho-metalated. This species is shown
in Scheme 1 and exhibits T1(min) values for its three hydride
resonances of 0.77, 0.77, and 0.51 s, respectively, at 248 K.
These values serve to confirm that 3 is a classical W(IV)
trihydride which explains why PHIP enhancement can be
observed in its hydride signals.20 The 183W signal of 3 appears
as a 31P-coupled multiplet at δ −2913 which proves to be close
to that of 2 thereby supporting their similar oxidation state and
18 electron counts.
(N2)2(dppe-κ2P)2 has previously been suggested to produce
W(dppe-κ2P)2 as the sole photoproduct.41 It subsequently
forms W(N2)(dppe-κ2P)2 and then W(N2)2(dppe-κ2P)2 upon
back-reaction with N2, although 15N labeling suggests a similar
sequential pathway for the original ligand loss process.46 The
first of the studies reported here was completed using in situ
irradiation of a sample of 1 under an atmosphere of H2 within
the NMR system at 233 K using a 325 nm HeCd laser.
One dominant photoproduct, 3 was evident in the resulting
NMR spectra, and with para-H2, the associated 1H NMR
spectra contain the same three hydride resonances all of which
exhibited a strong PHIP effect. On the basis of the intensities of
the PHIP enhanced signals, we conclude that H2 addition to
the sites giving rise to the δ −2.69 and δ −2.93 signals is
dominant. Under these conditions, a weakly polarized hydride
signal for 2 was also seen (Figure 2).
Conversion of WH3{Ph(C6H4)PCH2CH2PPh2-κ2P}(dppe-
κ2P) (3) to W(H)4(dppe-κ2P)2 (2). The slow conversion of 3
into the tetrahydride complex 2 was found to occur upon
warming a solution of 3 in toluene-d8 in the presence of 3 atm
H2. However, a new hydride resonance due to a further species,
4, was also detected at δ −2.4 (Figure 3a) in these experiments,
although at a level of less than 5% conversion. The conversion
of 3 to 2 requires the reformation of the activated CH bond
which in turn offers an explanation for the observation of PHIP
in the ortho-phenyl proton resonance of 2 at δ 7.42 as described
earlier; a para-H2 derived proton is placed in this site through
this exchange process.
Characterization of WH3{Ph(C6H4)PCH2CH2Ph2-κ2P}-
(dppe-κ2P) (3). External UV irradiation of a similar sample
at 223 K using a 180 W Oriel UV lamp was then used to
produce an NMR sample that was found to contain essentially
1
3 (Figure 1b illustrates the hydride region of the resulting H
Kinetic Studies on the Conversion of 3 to 2. When this
conversion was monitored in THF-d8 under 3 atm of H2, only 2
is formed with a rate constant for conversion at 293 K of 4.2 ×
10−5 s−1. This reaction was then monitored over the
NMR spectrum). This enabled the complete low temperature
NMR characterization of 3 via COSY, NOE and HMQC
methods. Notably, the three hydride signals of 3 now clearly
18260
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
conclusively, and while a W(H)2(dppe-κ2P)2 formulation would
match with the spectroscopic data, the DFT study now
described suggests that this is not the case.
Mapping of the Reaction Pathway of W(N2)2(dppe-
κ2P)2 (1) with Hydrogen. To rationalize these reactions
further, a series of DFT calculations was performed on the
model system W(N2)2((Ph)HPCH2CH2PH2-κ2P)-
(H2PCH2CH2PH2-κ2P) (1m) for which the dppe ligands has
been replaced by the simplified phosphine ligands indicated. An
optimized DFT structure for 1m was computed using implicit
solvation for THF. The structural features of 1m correspond
well with those of 1 as detailed in the Supporting Information.
This structure was used as a starting point in a subsequent
geometry optimization calculation of the bis-N2 loss-product
W((Ph)HPCH2CH2PH2-κ2P)(H2PCH2CH2PH2-κ2P) (6m).
The most stable geometry exists as a triplet ground state
6m3, which lies 210 kJ mol−1 above 1m in Gibbs Free Energy
terms (Scheme 2, further reactivity, however, is via 6m1, the
Figure 3. (a) 1H hydride ligand NMR signal for W(H)6(dppe-
κ2P)(dppe-κ1P) (4) and (b) three 31P decoupled PHIP polarized
hydride signals seen for W(H)3(OH)(dppe-κ2P) (5).
temperature range 283−323 K and ΔH⧧ and ΔS⧧ values of 79
3 kJ mol−1 and −54 11 J K−1 mol−1, respectively, were
determined from the associated rate data. These values are
consistent with a transition state that involves a reduction in the
number of species present and hence a contribution from H2
binding in the rate limiting step. Furthermore, when this
reaction is completed with D2 rather than H2 at 293 K, while
there is no evidence for D incorporation into 3, kH/kD is 1.09
which supports the involvement of H2 in this pathway. The
corresponding values of ΔH⧧ and ΔS⧧ for the reaction of 3
with D2 are 70.6 10 kJ mol−1 and −85 32 J K−1 mol−1,
respectively. The value of ΔG⧧ (300) for the reaction with D2
was determined as 96.2 kJ mol−1 compared with that of 95.0 kJ
mol−1 for reaction with H2 which confirms that an interaction
with the incoming ligand helps lower the barrier to this
reaction. The rate data that are used to determine these values
is presented in Table 3. The associated Eyring plot, and a
Scheme 2. DFT Reaction Coordinate Profile for the
Conversion of 1m into 2m without Specific THF
Involvement
Table 3. Rate Data for the Conversion of 3 into 2 through
Reaction with H2 and D2, Respectively
temperature/K
kH/s
kD/s
283
293
303
313
323
9.75 × 10−6
4.19 × 10−5
1.27 × 10−4
2.98 × 10−4
9.73 × 10−4
9.29 × 10−6
3.84 × 10−5
5.08 × 10−5
2.40 × 10−4
4.39 × 10−4
singlet). Oxidative addition of H2 then proceeds to form the
classical dihydride complex 7m in a barrierless reaction
involving the end-on approach of H2. Similar behavior has
been reported in other systems.48
typical kinetic trace are presented in Figure 4. The kinetic data
that are presented in Figure 4 was recorded at 293 K and
corresponds to monitoring the conversion of complex 3 into 2.
The addition of a second molecule of H2 to 7m then forms
the expected tetrahydride product W(H)4((Ph)-
HPCH2CH2PH2-κ2P)(H2PCH2CH2PH2-κ2P) (2m) which lies
25 kJ mol−1 below 1m in a barrierless reaction. Significantly, the
formation of W(H)3{(C6H4)HPCH2CH2PH2-κ2P})-
(H2PCH2CH2P(H)2-κ2P) (3m) occurs via ortho-metalation of
7m via a 26 kJ mol−1 transition state. 3m is, however, 33 kJ
mol−1 lower in energy than 7m and therefore corresponds to a
thermodynamically viable product in accordance with the
experimental observation of 3. 3m can also be formed in a
reaction sequence that first involves CH activation (to form
8m) from 6m1, where the barrier to ortho-metalation is now 20
kJ mol−1, and then H2 addition.
Figure 4. (a) Eyring plot for the conversion of 3 into 2; (b) kinetic
trace showing how the proportions of 3 ( ) and 2 ( ) change with
reaction time (sec) at 293 K.
■
●
However, the explicit solvation of 6m3 with THF yields the
complex W((Ph)HPCH2CH2PH2-κ2P)(H2PCH2CH2PH2-
κ2P)(THF)2 which exists in its most stable form as a singlet
(6s1), 88 kJ mol−1 more stable than the triplet, 6s3, and just 3 kJ
mol−1 less stable than 6m3 (Scheme 3). Given that the
experimental reaction is carried out in THF, it is reasonable to
assume that 6s1 exists in solution. Following loss of a single
THF ligand from 6s1, oxidative addition of H2 proceeds to form
It should be noted that in the absence of H2, 3 decomposes,
while also forming a new hydride-containing species. This
complex is highly fluxional but decomposes on exposure to H2
rather than forming 2 or 3. At 263 K, two mutually coupled
hydride resonances are visible at δ −0.82 and −1.03. The
corresponding 183W signal of this species appears at δ −2245.
Despite our best efforts, we have not been able to identify it
18261
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
shown in Figure 3a. This hydride resonance also possesses 183
satellites with JHW = 28 Hz. The three 31P nuclei which couple
to the hydride produced signals at δ 54.7 (d) and 42.3 (td),
while a third signal for an uncoordinated moiety is evident as a
doublet with JPP = 38 Hz at δ −12.4; this resonance showed no
coupling to the hydride ligands. The T1 value of 0.89 s for the
hydride ligands in 4, at 228 K, is consistent with a high
oxidation state and therefore classical hydride binding.20 4 must
therefore be the 18 electron complex W(H)6(dppe-κ2P)(dppe-
κ1P).
W
Scheme 3. DFT Reaction Coordinate Profile for the
Conversion of 1m into 2m with Specific THF Involvement
These conclusions are supported by the fact that the known
nine coordinate hexahydride complex W(H)6(PMe2Ph)3 has
similar JPH (36 Hz) and JWH (27 Hz) to those observed for 4.51
These data are therefore fully consistent with the proposed
W(VI) hexahydride formulation of 4 and we note the
diagnostically low value of JWH of 28 Hz compared with that
of 2 (38 Hz).51
The photochemical procedure described above reached a
photostationary state with 2 and 4 present in a 1:1 ratio and
with no evidence for 3 found in the associated NMR spectra.
Furthermore, there was no evidence of any net reaction when 2
was irradiated in the absence of H2. Repetition of the reaction
of 2 with para-H2, failed to produce PHIP in the hydride signals
observed for either 2 or 4. These results therefore suggest that
UV irradiation of 2 under these conditions leads to phosphine
dissociation rather than the H2 loss suggested by others
previously.52
the classical dihydride complex 7s in a barrierless reaction
involving the end-on approach of H2. In this scenario, THF
clearly plays an important role in stabilizing the dihydride
complex W(H)2((Ph)HPCH2CH2PH2-κ2P)(H2PCH2CH2PH2-
κ2P)(THF) (7s) over its unsolvated 16-electron counterpart
7m.
These DFT results are consistent with the experimental
observation of PHIP in the hydride resonances of both 2 and 3
when explicit THF solvation is included, and confirm that there
is little contribution from any W-η2-H2 type species on the
reaction coordinate which connects 1 with 2. We note that, in
support of this deduction, Ru(dppe-κ2P)(CO)2 does indeed
add H2 via the electronic singlet intermediate, with PHIP being
detected in the resulting hydride NMR signals.49 In contrast,
the analogous intermediate Fe(dppe-κ2P)(CO)2 has been
shown by DFT to exist as a triplet which is consistent with
the failure to observe PHIP for Fe(H)2(dppe-k2P)(CO)2.49
An alternative mechanism involving loss of a single N2 from
1m, yields the 16-electron complex W(N2)((Ph)-
HPCH2CH2PH2-κ2P)(H2PCH2CH2PH2-κ2P) for which a sim-
ilar barrier exists to ortho-metalation as that determined for the
equivalent THF complex, W((Ph)HPCH2CH2PH2-κ2P)-
(H2PCH2CH2PH2-κ2P) (THF). However, formation of the
N2 ortho-metalation product is thermodynamically unfavorable
with respect to W(N2)((Ph)HPCH2CH2PH2-κ2P)-
(H2PCH2CH2PH2-κ2P)(THF). Computed binding enthalpies
for the first and second N2 ligands of 1m are 214 and 129 kJ
mol−1, respectively. This is consistent with the experimental
observation that the rate of binding of the second N2 ligand
occurs much slower than that of the first.50 Furthermore, when
coupled with the relative thermodynamic instability of the N2
ortho-metalation product, an explanation emerges for why no
net photochemistry is observed for 1 in the absence of H2.
Formation of W(H)6(dppe-κ2P)(dppe-κ1P) (4) by UV
Irradiation of W(H)4(dppe-κ2P)2 (2). The chemistry
discussed so far is further complicated by the fact that UV
irradiation of a preformed sample containing 3 under H2 at 295
K leads to the detection of both 2 and 4. In contrast, irradiation
of 1 with H2 at 295 K results in the dominant photoproduct 2
with much lower levels of 3 and 4. All four of these species are
therefore photochemically active under these conditions.
When 2 is irradiated under an H2 atmosphere at 263 K, the
dominant product is 4. Notably, the single hydride signal seen
for 4 appears at δ −2.4 and couples to only three 31P nuclei
with JPH = 26 Hz (triplet) and 36 Hz (doublet), respectively, as
Additionally, when a sample of 4 is warmed with para-H2, no
PHIP is evident in the detected hydride signals of 2. This
observation confirms that the displacement of H2 and
conversion of the dppe-κ1P ligand to its κ2 form is
intramolecular with respect to four of the six hydride ligands
of 4.
Formation of W(H)3(OH)(dppe-κ2P)2 (5). In a number of
these reactions, a further product was often observed but only if
NMR solvents were used without drying.
When a wet toluene-d8 solution of 1 was left overnight, slow
decomposition occurs such that 31P NMR signals for 2, free
dppe, the mono phosphine oxide of dppe and a new complex 5
can be detected. In the 1H NMR spectrum, three new high-field
signals for 5 dominate at δ −2.26, δ −2.37 and δ −3.00. These
signals, which occur in the hydride region, appear even though
no H2 has been explicitly added to the sample.
However, upon addition of para-H2 to the same sample, the
hydride signals at δ −2.26, δ −2.37, as well as a previously
masked peak at δ 1.59 (with a peak envelope of ca. 200 Hz
which indicates the presence of several strong couplings to 31P)
all show PHIP polarization (Figure 3b). The signal at δ −3.00
1
which possesses much smaller H−31P splittings is apparently
unaffected by the addition of para-H2 and when D2O and para-
H2 were used as reagents, only the two signals at δ −2.26, δ
−2.37 were polarized.
1
In an H2O-containing sample, the same H NMR integral
values were obtained for the three high field signals, the eight
ethane bridge protons, and the resonance at δ 1.59.
Significantly, COSY spectroscopy revealed that the δ 1.59
signal of 5 couples to the two hydride signals δ −2.26 and δ
−2.37 while EXSY measurements confirm that the hydrogen
atoms in these three sites undergo positional interchange. The
resonance at δ −3.00 is not involved in this process. In the
1
corresponding H−31P HMQC measurement, values of JPH
=
100, 30, and 60 Hz produce strong connections to the δ 1.59
1H NMR signal which is therefore confirmed as a W-hydride
18262
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
resonance. The three high field resonances have T1 values of
0.28, 0.28, and 0.74 s, respectively, at 268 K. Complex 5 yields a
further 40 aromatic proton resonances that arise from eight
distinct and intact phenyl rings. It also gives four 31P NMR
signals for tungsten-coordinated phosphorus centers that
appear at δ 36.6, 45.5, 47.3 and 64.0.
Scheme 4. Conversion of 1 to 2 Where H2O Can Act as the
Source of H2
a
The identity of 5 was secured through a combination of
EXSY measurements that probed the behavior of the site
yielding the δ 3.00 signal together with H2O doping studies
that were conducted in toluene-d8 solution. For the doping
studies, a sample of 1 was prepared and a 20-fold excess of dppe
added prior to the addition of 5 μL of H2O. Over a period of
several days at room temperature, the conversion of 1 into 5
was observed. On the basis of 31P signal integrations, 50% of 1
converted into 5 and a similar amount of PPh2CH2CH2Ph2P-
(O). At this stage no H2 was evident in solution. Concurrent
UV irradiation increased the rate of reaction while generating
mainly 2 and 4 and free H2; the addition of H2O to a solution
of 3 produced a mixture of 2 and 5.
a
These transformations can be initiated thermally and photochemi-
cally.
Table 4. Diffusion Data for 2, 3, 4, and 5
complex, solvent, and
temperature
diffusion coefficient
resonance /δ
(cm2/s)
EXSY measurements revealed that magnetization transfer
occurs from the protons giving rise to the δ −3.00 signal of 5 to
those associated with the peak at δ 0.47, due to H2O. This
process was enhanced by increasing the water concentration.
Collectively, these observations allow 5 to be identified as
W(H)3(OH)(dppe-κ2P)2 where the rate data associated with
the proton transfer step yield ΔH⧧ = 42 5 kJ mol−1 and ΔS⧧
= −90.5 18 J K−1 mol−1. The protons giving rise to the δ 1.59
hydride signal also show exchange with H2O in the EXSY data.
In addition, the three hydride signals show evidence for
positional exchange.
2, THF, 293 K
3, THF, 293 K
−3.94
−0.25
−2.98
−3.12
−2.85
−2.29
−2.40
−3.00
1.27
1.50
1.50
1.50
2.42
0.57
0.57
0.57
4, THF, 293 K
5, Toluene, 273 K
the related complex W(H)3(η2-C6H4PMe2)(PMe2Ph)3 from
56
t
In support of the assignment of 5 as a tungsten hydroxide
complex, it should be noted that the hydroxyl proton resonance
of the related complex Cp*(PMe3)Ir(Ph)OH appears as a
broad singlet at δ −3.1 while that of Ru(IMes)2(CO)2(OH)H
appears at δ −3.75.53,54 In contrast, the aquo complex
W(CO)3(PiPr3)(H2O) yields a signal for the coordinated
H2O at δ 3.03.55 Consequently, the δ −3.00 signal of 5 is
consistent with that of a metal hydroxyl. This is further
supported by the fact that the addition of D2O results in the
rapid loss of the W(OH) resonance of 5 with subsequent
reaction with para-H2, yielding only the PHIP enhanced δ
−2.26 and δ −2.37 signals.
On a slower time scale, 5 converts thermally into 2 although
ultimately in the presence of an excess of water only phosphine
oxides of dppe remain according to 31P NMR spectroscopy.
This is consistent with the slow reductive elimination of H2O
from the metal center in 5 and subsequent reaction to form 2
or 3 as shown in Scheme 4. These observations also reveal that
the 16 electron fragment W(H)2(dppe-κ2P)2 is capable of
intramolecular CH bond activation, in agreement with the DFT
study discussed earlier.
WH2Cl2(PMe2Ph)4 via reaction with BuLi. Furthermore,
Hidai et al. have speculated that warming benzene solutions of
1 to reflux leads to the formation of meso-o-
C6H4(PPhCH2CH2PPh2)2 via the proposed reaction inter-
mediate [WH{Ph(C6H4)PCH2CH2PPh2-κ2P}(dppe-κ2P)]; we
note that some of their reported NMR data10 appear to match
that presented here for 3. The observation of a CH bond
activation pathway during the formation of 2 is therefore
entirely consistent with the known chemistry of related
systems.
The theoretical studies used W(N2)2((Ph)HPCH2CH2PH2-
κ2P)(H2PCH2CH2PH2-κ2P) (1m) as a starting point from
which to explore these reactions. They revealed that N2 loss
leads to W(N2)2((Ph)HPCH2CH2PH2-κ2P)(H2PCH2CH2PH2-
κ2P) (6m3) which exists as a triplet unless explicitly solvated by
two molecules of THF. ortho-Metalation and H2 addition were
both shown to be feasible reaction steps prior to the formation
of 3 and 2, respectively. These results therefore rationalized the
observation of both 3 and 2 in the reaction chemistry of 1.
They also help explain the detection of PHIP in the
corresponding reaction products.
Diffusion Ordered Spectroscopy. To confirm that these
signals all originate from molecules with similar masses, a series
of DOSY measurements were undertaken. These resulted in
similar diffusion coefficients for 2, 3, 4, and 5, thereby
confirming our mononuclear hypothesis, as illustrated in Table
4.
The trapping of hydrogen by organometallic complexes and
the reverse process have proven important in energy storage
applications.57 It has been demonstrated in this work that
W(N2)2(dppe-κ2P)2 (2) is capable of releasing H2 from H2O in
conjunction with the formation of a phosphine oxide and
ultimately W(H)4(dppe-κ2P)2 (2). This transformation involves
the metal hydroxyl-hydride complex W(H)3(OH)(dppe-κ2P)2
(5) via the reaction sequence illustrated in Scheme 4. Such
species have been shown to play a role in organic oxidations.58
This reaction therefore illustrates a route to H2 production
from H2O facilitated, in this case by phosphorus acting as a
suitable oxygen acceptor.
CONCLUSIONS
■
The conversion of W(N2)2(dppe-κ2P)2 (1) into W(H)4(dppe-
κ2P)2 (2) has been shown to involve the initial formation of the
ortho-metalated CH activation product [WH3{Ph(C6H4)-
PCH2CH2PPh2-κ2P}(dppe-κ2P)] (3). Caulton et al. prepared
18263
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
(7) Bell, B.; Chatt, J.; Leigh, G. J. J. Chem. Soc. D 1970, 842.
(8) Ishida, T.; Mizobe, Y.; Tanase, T.; Hidai, M. J. Organomet. Chem.
1991, 409, 355.
(9) Seino, H.; Arita, C.; Nonokawa, D.; Nakamura, G.; Harada, Y.;
Mizobe, Y.; Hidai, M. Organometallics 1999, 18, 4165.
(10) Arita, C.; Seino, H.; Mizobe, Y.; Hidai, M. Bull. Chem. Soc. Jpn.
2001, 74, 561.
(11) Sattler, A.; Parkin, G. Nature 2010, 463, 523.
(12) Chatt, J.; Pearman, A. J.; Richards, R. L. J. Chem. Soc., Dalton
Trans. 1977, 2139.
When these reactions are conducted with para-H2, PHIP is
evident in the hydride signals of 2, 3 and 5. This confirms the
utility of PHIP as a powerful method for examining the
reactivity of polyhydride complexes of this type. Furthermore,
the experimental and computation results also suggest that
nonclassical dihydride complexes do not play a significant role
in the oxidative addition process. In contrast, H2 addition to
W(CO)3(PCy3)2 was shown not to produce detectable PHIP in
the W(H)2(CO)3(PCy3)2 product. In this latter case, a role for
a nonclassical dihydrogen complex in the addition process has
been well established.3−5 The postulation here that 3 and 4
exist as classical W(IV) and W(VI) species was further
supported by the long T1(min) values of the hydride ligands
(>0.3 s), the observation of appropriate JHW couplings and the
detection of 183W signals for 2, 3 and 4. The 183W resonance for
W(VI) 4 appears at the high field value of δ 4000, in
accordance with its high oxidation state. This compares with
values of δ 3363 for 2 and δ 2913 for 3 both of which
complexes exists as W(IV) species. Attempts to locate the
corresponding signal in 5 failed, presumably as a consequence
of its dynamic behavior leading to signal broadening.
(13) Jia, G.; Morris, R. H.; Schweitzer, C. T. Inorg. Chem. 1991, 30,
593.
(14) Nishibayashi, Y.; Iwai, S.; Hidai, M. Science 1998, 279, 540.
(15) Nishibayashi, Y.; Takemoto, S.; Iwai, S.; Hidai, M. Inorg. Chem.
2000, 39, 5946.
(16) Hlatky, G. G.; Crabtree, R. H. Coord. Chem. Rev. 1985, 65, 1.
(17) Crabtree, R. H.; Lavin, M. J. Chem. Soc., Chem. Commun. 1985,
1661.
(18) Hamilton, D. G.; Crabtree, R. H. J. Am. Chem. Soc. 1988, 110,
4126.
(19) Ammann, C.; Isaia, F.; Pregosin, P. S. Magn. Reson. Chem. 1988,
26, 236.
(20) Michos, D.; Luo, X. L.; Faller, J. W.; Crabtree, R. H. Inorg.
It has previously been reported that UV irradiation of
W(H)4(dppe-κ2P)2 under N2 leads to low yields of ammonia59
and alkene hydrogenation. It has also been reported that H2
loss corresponds to the main photochemical process in such
systems.50,52 Here we have demonstrated that W(H)6(dppe-
κ2P)(dppe-κ1P) (4) is formed under UV irradiation which
suggests that 16-electron W(H)4(dppe-κ2P)(dppe-κ1P) and not
W(H)2(dppe-κ2P)2 is responsible for these conversions.
Thus, the results presented here demonstrate that CH bond
activation, phosphine dechelation, and the oxidative addition of
OH bonds are all important in the chemistry of W(N2)2(dppe-
κ2P)2 (1).
Chem. 1993, 32, 1370.
(21) Oglieve, K. E.; Henderson, R. A. J. Chem. Soc., Chem. Commun.
1992, 441.
(22) Muckerman, J. T.; Fujita, E.; Hoff, C. D.; Kubas, G. J. J. Phys.
Chem. B 2007, 111, 6815.
(23) Li, J.; Yoshizawa, K. Chem.Eur. J. 2012, 18, 783.
(24) Maseras, F.; Lledos, A.; Clot, E.; Eisenstein, O. Chem. Rev. 2000,
100, 601.
(25) Balu, P.; Baskaran, S.; Kannappan, V.; Sivasankar, C. Polyhedron
2012, 31, 676.
(26) Sandhya, K. S.; Suresh, C. H. Dalton Trans. 2012, 41, 11018.
(27) 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.
(28) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1999, 121, 6336.
(29) Ball, G. E.; Darwish, T. A.; Geftakis, S.; George, M. W.; Lawes,
D. J.; Portius, P.; Rourke, J. P. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
1853.
(30) Cunningham, J. L.; Duckett, S. B. Dalton Trans. 2005, 744.
(31) Matthews, S. L.; Heinekey, D. M. J. Am. Chem. Soc. 2006, 128,
2615.
ASSOCIATED CONTENT
* Supporting Information
Characterization of 3 and 5, and the DFT derived energetics
and coordinates for all the species discussed. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
sbd3@york.ac.uk
Notes
(32) Lyon, C. E.; Suh, E. S.; Dobson, C. M.; Hore, P. J. J. Am. Chem.
Soc. 2002, 124, 13018.
■
(33) Blazina, D.; Dunne, J. P.; Aiken, S.; Duckett, S. B.; Elkington, C.;
McGrady, J. E.; Poli, R.; Walton, S. J.; Anwar, M. S.; Jones, J. A.;
Carteret, H. A. Dalton Trans. 2006, 2072.
(34) Dunne, J. P.; Blazina, D.; Aiken, S.; Carteret, H. A.; Duckett, S.
B.; Jones, J. A.; Poli, R.; Whitwood, A. C. Dalton Trans. 2004, 3616.
(35) Anwar, M. S.; Blazina, D.; Carteret, H. A.; Duckett, S. B.;
Halstead, T. K.; Jones, J. A.; Kozak, C. M.; Taylor, R. J. K. Phys. Rev.
Lett. 2004, 93, No. 040501.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful for financial support from the BBSRC (P.J.C.),
AstraZeneca and the Spanish MEC Consolider Ingenio 2010-
ORFEO-CSD2007-00006 (B.E.) research programme. We also
thank Professor Robin Perutz for helpful discussions.
(36) Duckett, S. B.; Mewis, R. E. Acc. Chem. Res. 2012, 45, 1247.
(37) Benn, R.; Brenneke, H.; Heck, J.; Rufinska, A. Inorg. Chem.
1987, 26, 2826.
(38) Carlton, L.; Emdin, A.; Lemmerer, A.; Fernandes, M. A. Magn.
Reson. Chem. 2008, 46, S56.
REFERENCES
■
(39) Crabtree, R. H.; Hlatky, G. G. Inorg. Chem. 1982, 21, 1273.
(40) Thomas, A.; Haake, M.; Grevels, F. W.; Bargon, J. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 755.
(41) Thomas, R. J. W.; Laurence, G. S.; Diamantis, A. A. Inorg. Chim.
Acta 1978, 30, L353.
(1) Bell, B.; Chatt, J.; Leigh, G. J. J. Chem. Soc., Dalton Trans. 1972,
2492.
(2) Hidai, M.; Mizobe, Y. Chem. Rev. 1995, 95, 1115.
(3) Kubas, G. J.; Unkefer, C. J.; Swanson, B. I.; Fukushima, E. J. Am.
Chem. Soc. 1986, 108, 7000.
(42) Carlton, L.; Emdin, A.; Lemmerer, A.; Fernandes, M. A. Magn.
Reson. Chem. 2008, 46, S56.
(43) Schott, D.; Sleigh, C. J.; Lowe, J. P.; Duckett, S. B.; Mawby, R. J.;
Partridge, M. G. Inorg. Chem. 2002, 41, 2960.
(44) Duckett, S. B.; Wood, N. J. Coord. Chem. Rev. 2008, 252, 2278.
(4) Khalsa, G. R. K.; Kubas, G. J.; Unkefer, C. J.; Vandersluys, L. S.;
Kubatmartin, K. A. J. Am. Chem. Soc. 1990, 112, 3855.
(5) Kubas, G. J.; Ryan, R. R.; Swanson, B. I.; Vergamini, P. J.;
Wasserman, H. J. J. Am. Chem. Soc. 1984, 106, 451.
(6) Heinekey, D. M.; Oldham, W. J. Chem. Rev. 1993, 93, 913.
18264
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265

Journal of the American Chemical Society
Article
(45) Desrosiers, P. J.; Cai, L. H.; Lin, Z. R.; Richards, R.; Halpern, J. J.
Am. Chem. Soc. 1991, 113, 4173.
(46) Archer, L. J.; George, T. A. Inorg. Chim. Acta, Lett. 1980, 44,
L129.
(47) Mohr, F.; Priver, S. H.; Bhargava, S. K.; Bennett, M. A. Coord.
Chem. Rev. 2006, 250, 1851.
(48) Macgregor, S. A.; Eisenstein, O.; Whittlesey, M. K.; Perutz, R.
N. J. Chem. Soc., Dalton Trans. 1998, 291.
(49) Schott, D.; Callaghan, P.; Dunne, J.; Duckett, S. B.; Godard, C.;
Goicoechea, J. M.; Harvey, J. N.; Lowe, J. P.; Mawby, R. J.; Muller, G.;
Perutz, R. N.; Poli, R.; Whittlesey, M. K. Dalton Trans. 2004, 3218.
(50) Pivovarov, A. P. Russ. Chem. Bull. 2008, 57, 289.
(51) Crabtree, R. H.; Hlatky, G. G. Inorg. Chem. 1984, 23, 2388.
(52) Graff, J. L.; Sobieralski, T. J.; Wrighton, M. S.; Geoffroy, G. L. J.
Am. Chem. Soc. 1982, 104, 7526.
(53) Woerpel, K. A.; Bergman, R. G. J. Am. Chem. Soc. 1993, 115,
7888.
(54) Jazzar, R. F. R.; Bhatia, P. H.; Mahon, M. F.; Whittlesey, M. K.
Organometallics 2003, 22, 670.
(55) Kubas, G. J.; Burns, C. J.; Khalsa, G. R. K.; Vandersluys, L. S.;
Kiss, G.; Hoff, C. D. Organometallics 1992, 11, 3390.
(56) Rothfuss, H.; Huffman, J. C.; Caulton, K. G. Inorg. Chem. 1994,
33, 2946.
(57) Kubas, G. J. J. Organomet. Chem. 2009, 694, 2648.
(58) Bryndza, H. E.; Tam, W. Chem. Rev. 1988, 88, 1163.
(59) Dziegielewski, J. O.; Gilbortnowska, R.; Mrzigod, J.; Malecki, J.
G. Polyhedron 1995, 14, 1375.
18265
dx.doi.org/10.1021/ja302202q | J. Am. Chem. Soc. 2012, 134, 18257−18265