trans-W(Cmesityl)(dmpe)2H: Revealing a highly polar W-H bond and H-mobility in liquid and solid state

Article abstract of DOI:10.1021/ja069140k

The isotopomeric complexes trans-W(Cmesityl)[(C(H,D)₃) ₂PCH₂CH₂P(C(H,D)₃)₂] ₂(H,D) 1 -4 were prepared. 2 (W(C_meSityl)(dmpe) ₂D) was used to study the Deuterium Quadrupole Coupling Constant (DQCC) and the ionicity of the W-D bond (DQCC = 34.1 kHz; ionicity 85%). 1 (W(C_mesityl)(dmpe)₂H) shows several dynamic exchange processes in solution, such as H_w/H_w, H _w/ortho-Me_mesityl, and H_w/H₂ exchanges observed by NMR in combination with deuterium labeling studies and double label crossover experiments. Except for the H_w/H₂, these reactions comprise elementary steps, which also appear along the isomerization pathway of 1 into (2,3,5-trimethylphenylcarbyne)(dmpe) ₂WH (5) at 60 °C. 5 was characterized by an X-ray diffraction study. In the solid state only an H_w/Me_p exchange process prevails appearing at higher temperatures, which was identified by NMR and by Quasielastic Neutron Scattering. The latter also provided an activation barrier of 5 kcal/mol and a jump width for the moving H nucleus in agreement with the H_W...Me_p distance of the X-ray diffraction study of 1.

Full text of DOI:10.1021/ja069140k

Published on Web 05/12/2007
trans-W(Cmesityl)(dmpe)2H: Revealing a Highly Polar W-H
Bond and H-Mobility in Liquid and Solid State
Fenglou Zou,† Franck Furno,† Thomas Fox,† Helmut W. Schmalle,† Heinz Berke,*,†
,|
Ju¨rgen Eckert,*,‡,§ Zema Chowdhury,| and Peter Burger‡,
Contribution from the Anorganisch-chemisches Institut, UniVersita¨t Zu¨rich, Winterthurerstrasse
190, CH-8057 Zu¨rich, Switzerland, LANSCE-LC, Los Alamos National Laboratory, Los Alamos,
New Mexico 87545, Materials Research Laboratory, UniVersity of California, Santa Barbara,
California 93106, and NIST Center for Neutron Research, Bldg 235,
Gaithersburg, Maryland 20899
Received January 10, 2007
Abstract: The isotopomeric complexes trans-W(Cmesityl)[(C(H,D)3)2PCH2CH2P(C(H,D)3)2]2(H,D) 1-4 were
prepared. 2 (W(Cmesityl)(dmpe)2D) was used to study the Deuterium Quadrupole Coupling Constant (DQCC)
and the ionicity of the W-D bond (DQCC ) 34.1 kHz; ionicity 85%). 1 (W(Cmesityl)(dmpe)2H) shows several
dynamic exchange processes in solution, such as HW/HW, HW/ortho-Memesityl, and HW/H2 exchanges observed
by NMR in combination with deuterium labeling studies and double label crossover experiments. Except
for the HW/H2, these reactions comprise elementary steps, which also appear along the isomerization
pathway of 1 into (2,3,5-trimethylphenylcarbyne)(dmpe)2WH (5) at 60 °C. 5 was characterized by an X-ray
diffraction study. In the solid state only an HW/Mep exchange process prevails appearing at higher
temperatures, which was identified by NMR and by Quasielastic Neutron Scattering. The latter also provided
an activation barrier of 5 kcal/mol and a “jump width” for the moving H nucleus in agreement with the
HW‚‚‚Mep distance of the X-ray diffraction study of 1.
Transition metal hydride (TMH) complexes1 are viewed as
the key components in a variety of organometallic transforma-
tions and catalytic schemes, during which they insert a wide
range of unsaturated compounds into the TMH bonds.2 In
mechanistic terms many of these insertion steps are not,
however, insertions in the sense of intramolecular H shifts at
the metal center but rather comprise in their intimate part hydride
transfers to appropriate electrophiles.3 This may even be the
preferred route of TMH complexes with enhanced hydride donor
abilities of the TM-H bond.4 However, a free H- ion has never
been observed in such reactions or detected in solution, so that
any research finding to this end would be of considerable
importance.
We have recently prepared various TMH complexes with
strong hydridic polarizations in the TMH bonds. One of these
compounds is trans-W(CMes)(dmpe)2H (dmpe ) 1,2-bis-
(dimethylphosphino)ethane) 1.5 The synthesis, full characteriza-
† Universita¨t Zu¨rich.
‡ Los Alamos National Laboratory.
§ University of California.
| NIST Center for Neutron Research.
Present address: Institut fu¨r Anorganische Chemie, Universita¨t Ham-
burg, Martin-Luther-King-Platz 6, D-20146 Hamburg, Germany.
(1) Kubas, G. J. Metal Dihydrogen and σ-Bond Complexes; Kluwer, Academic/
Plenum: New York, 2001.
(2) For example, see: (a) Houben-Weyl Methoden der Organischen Chemie;
Georg Thieme Verlag: Stuttgart, Germany, 1981; Vol. IV/1d, pp 267-
282, 293-338; 1984; Vol. IV/1b, pp 141-288. (b) 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; Chapter 10. (c) Ojima, I.; Eguchi, M.; Tzamarioudaki, M. In
ComprehensiVe Organometallic Chemistry II; Abel, E. W., Stone, F. G.
A., Wilkinson, G., Eds.; Pergamon: Exeter, 1995; Vol. 12, Chapter 2. (d)
Berke, H.; Burger, P. Comments Inorg. Chem. 1994, 16, 279.
(3) (a) Gaus, P. L.; Kao, S. C.; Youngdahl, K.; Darensbourg, M. Y. J. Am.
Chem. Soc. 1985, 107, 2428. (b) Collman, J. P.; Wagenknecht, P. S.; Lewis,
N. S. J. Am. Chem. Soc. 1992, 114, 5665. (c) Hembre, R. T.; McQueen, S.
J. Am. Chem. Soc. 1994, 116, 2141. (d) Bullock, R. M.; Song, J.-S. J. Am.
Chem. Soc. 1994, 116, 8602. (e) Luan, L.; Song, J.-S.; Bullock, R. M. J.
Org. Chem. 1995, 60, 7170. (f) Voges, M. H.; Bullock, R. M. J. Chem.
Soc., Dalton Trans. 2002, 759. (g) Song, J.-S.; Szaldo, D. J.; Bullock, R.
M. Inorg. Chim. Acta 1997, 259, 161. (h) Cheng, T. Y.; Bullock, R. M.
Organometallics 2002, 21, 2325.
tion, and insertion type reactivity of 1 toward small unsaturated
organic and inorganic molecules have been reported elsewhere.5
These studies, together with density functional calculations,6
(4) Jacobsen, H.; Berke, H. Recent AdVances in Hydride Chemistry; Poli, R.,
Ed.; Elsevier: Amsterdam, 2001; p 89.
(5) (a) Furno, F.; Fox, T.; Schmalle, H. W.; Berke, H. Organometallics 2000,
19, 3620. (b) Furno, F.; Fox, T.; Alfonso, M.; Berke, H. Eur. J. Inorg.
Chem. 2001, 1559. (c) Furno, F.; Fox, T.; Schmalle, H. W.; Berke, H.
Chimia 1999, 53, 350.
(6) Jacobsen, H.; Berke, H. J. Chem. Soc., Dalton Trans. 2000, 3117.
9
10.1021/ja069140k CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 7195-7205
7195

A R T I C L E S
Zou et al.
Scheme 1
have indeed indicated an enhanced hydridic character, mainly
due to the design of the hydride complex bearing strong
σ-donating phosphines and a mesityl carbyne ligand in trans
position, which is known to exert a strong trans-influence/
effect.7 A X-ray diffraction analysis5a of 1 demonstrated that
the overall geometry about the tungsten center is that of a
slightly distorted octahedron with a distance for the W-H bond
of 2.0 Å, while DFT calculations revealed a W-H distance of
1.88 Å. Both results point to a long and relatively weak TM-H
bond.6
On the basis of these structural properties and the high activity
in H-transfer reactions, it was not too surprising to find a very
high value of the W-H bond ionicity determined by two
deuteride (T1 ) 300 s (298 K)) and did not indicate any
intermolecular exchange at this temperature.11,15
Isomerization of 1 into W[(C(TMP)](dmpe)2H (TMP )
2,3,5-Trimethylphenyl) (5). At 60 °C W(CMes)(dmpe)2H (1)
was observed to rearrange within 2 weeks to form W[(C(TMP)]-
(dmpe)2H (TMP ) 2,3,5-trimethylphenyl) (5) in benzene or
toluene solution with a 90% spectroscopic yield of 5 (isolated
yield 71%) (Scheme 1).
The isomerization was first thought to be initiated by light.
However, a UV irradiation experiment with a mercury lamp
(125 W) in a cold water bath did not reveal any changes.
The 1H NMR spectrum of 5 displays a quintet for the hydride
ligand at -5.62 ppm (2JPH ) 35 Hz, JWH ) 34 Hz) about 1
ppm at lower field than that of 1. The rearrangement of the
aromatic ring becomes also obvious in the 1H NMR spectrum,
since 5 shows for the MeAr groups three singlets of equal
intensity. In the aromatic chemical shift region two distinct
resonances are detected assigned to the two nonequivalent HAr
atoms. The 31P NMR spectrum indicated equivalence of all four
phosphorus nuclei showing a singlet at 25.6 ppm (JWP ) 276
Hz), which is shifted about 0.5 ppm to lower field in comparison
with 1.
The spectroscopically derived structure of 5 was eventually
confirmed by a single-crystal X-ray diffraction study. Red single
crystals were grown by cooling a hexane solution to -35 °C.
The ORTEP plot of 5 is shown in Figure 1. Selected bond
lengths and bond angles are listed in Table 1.
In the X-ray diffraction study the Hhydride ligand (H49) was
found in a difference Fourier map. Its coordinates were kept
fixed, and the isotropic displacement parameters were refined.
All other positions of H-atoms were recalculated after each
refinement cycle.
As shown in Figure 1, 5 possesses a similar structure to that
of 1 with a square bipyramidal geometry of the heavy atoms
around the tungsten center with four phosphorus atoms in the
quasi equatorial plane, while the Hhydride atom and the carbyne
ligand are located perpendicular to that plane in trans positions
similar to the structure of 1.
The structural data of 5 can be interpreted in terms of reduced
molecular strain in comparison with 1. For instance, as listed
in Table 1, the W(1)sC(13) bond length of 5 is 1.838(9) Å,
which is significantly shorter than the corresponding bond of 1
(1.863(6) Å). Presumably this shortening is due to less non-
bonding contacts between the MeAr and Medmpe groups (closest
distance of 5 2.519 Å). The presence of MeAr‚‚‚Medmpe steric
repulsion can also be derived from the C(14)sC(13)sW(1)
“hinging” angle of the aromatic ring of 5 (174.7(7)°) bending
2
different techniques. VT H-T1 NMR relaxation times were
determined in toluene-d8 solution for the deuterium labeled
compound 2, which were used for the calculation of the
deuterium quadrupole coupling constant (DQCC) and the ionic
bond character (i) in the covalent W-H bond.8 The rate of 2H
spin-lattice relaxation is dominated by quadrupolar interactions.
2
2 shows a broad multiplet at δ ) -6.60 ppm in its H NMR
spectrum. The logarithmic 2H-T1 times of the 2H resonance for
the D ligand of 2 measured at 213 K (76.7 MHz) and plotted
as a function of 1/T were used to extract a T1 min value of 59
ms, which corresponds to a DQCC of 34.1 kHz and (i) )
85%9,10 calculated from (i) ) 1-(DQCC/227).9 We have also
2
employed solid-state H NMR as an alternative technique to
evaluate the DQCC in 210 whereby the DQCC can be directly
extracted from the Pake doublet splitting.11 A similar value of
34.8 kHz for the DQCC was found in nice agreement with the
2
liquid H-T1 measurement. The degree of W-D bond ionicity
in 2 of about 85% is higher than values reported for (CO)5MnD
((i) ) 70%), Cp2WD2 ((i) ) 76%), WD(NO)(CO)2(PMe)2 ((i)
) 79%), Cp2ZrD2 ((i) ) 79%), WD(NO)(CO)(PMe)3 ((i) )
82%),9 and MoD(NO)(dmpe)2 ((i) ) 84%).12 The smallest
known experimental DQCC value, 33 kHz, (i) ) 85%, is that
of the LiD molecule,13 which is close to our results for the trans-
W(CMes)(dmpe)2D complex, which approaches the ionic limit
of an M-D bond with a zero DQCC value. On the other hand
a DQCC value of 227 kHz in HD14 represents the 100% covalent
end of the scale. The solid state measurements in addition
revealed a very long relaxation time for the metal bound
(7) Appleton, T. G.; Clark, H.; Manzer, L. Coord. Chem. ReV. 1973, 10, 335.
Atwood, J. D. Inorganic and Organometallic Reaction Mechanisms, 2nd
ed.; VCH Publishers: New York, 1997. Shriver, D. F., Atkins, P. W., Eds.
Inorganic Chemistry; W. H. Freeman and Co: New York, 1999. Basolo,
F.; Pearson, R. G. Progr. Inorg. Chem. 1962, 4, 381. Faller, J. W.; Lambert,
C.; Mazzieri, M. R. J. Organomet. Chem. 1990, 383, 161. Faller, J. W.;
Mazzieri, M. R.; Nguyen, J. T.; Parr, J.; Tokunaga, M. Pure Appl. Chem.
1964, 66, 1463.
(8) (a) Nietlispach, D.; Bakhmutov, V. I.; Berke, H. J. Am. Chem. Soc. 1993,
115, 9191. (b) Gusev, D. G.; Nietlispach, D.; Vymenits, A. B.; Bakhmutov,
V. I.; Berke, H. Inorg. Chem. 1993, 32, 3270. (c) Wharton, L.; Gold, L.
P.; Klemperer, W. J. Chem. Phys. 1962, 37, 2149.
(9) (a) Fee, F. W.; Chan, M. C. W.; Cheung, K. K.; Che, C. M. J. Organomet.
Chem. 1998, 563, 191. (b) Zhang, L.; Gamasa, M. P.; Gimeno, J.; Carbajo,
R. J.; Lopez-Ortiz, F.; Lanfranchi, M.; Tiripicchio, A. Organometallics
1996, 15, 4274.
(10) (a) Fyfe, C. A. Solid State NMR for Chemists; C.F.C. Press: Ontario, 1983.
(b) Dixon, M.; Overill, R. E.; Platt, E. J. Mol. Struct. 1978, 48, 115. (c)
Soda, G.; Chiba, T. J. Chem. Phys. 1969, 50, 439. (d) Butler, L. G.; Brown,
T. L. J. Am. Chem. Soc. 1981, 103, 6541. (e) Kim, A.; Fronczek, F. R.;
Butler, L. G.; Chen, S.; Keiter, E. A. J. Am. Chem. Soc. 1991, 113, 9090.
(11) Smith, J. A. S. J. Chem. Educ. 1971, 48, 39.
(12) Liang, F.; Schmalle, H. W.; Fox, T.; Berke, H. Organometallics 2003, 22,
3393.
(13) (a) Wei, Y.; Fung, B. M. J. Chem. Phys. 1971, 55, 1486. (b) Ireland, P. S.;
Olson, L. W.; Brown, T. L. J. Am. Chem. Soc. 1975, 97, 3548. (c) Jarrett,
W. L.; Farlee, R. D.; Butler, L. G. Inorg. Chem. 1987, 26, 1381.
(14) Guo, K.; Jarrett, W. L.; Butler, L. G. Inorg. Chem. 1987, 26, 3001.
(15) Hayashi, S. Inorg. Chem. 2002, 41, 2238. Hayashi, S. J. Phys. Chem. Solids
2003, 64, 2227. Hayashi, S. J. Solid State Chem. 2003, 170, 82. Hayashi,
S. J. Alloy Compd. 2002, 330, 443.
9
7196 J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007

trans-W(Cmesityl)(dmpe)2H: Polar W−H Bond and H-Mobility
A R T I C L E S
methyl groups are “turned on” presumably starting from an
agostic binding state as in the case of 1d. The first C-H
oxidative addition of Scheme 2 results in the cis-hydrido
methylene-bridged complex 1e. All the following oxidative
additions and their reversals C-H reductive eliminations are
assumed to be associated with a similar activation barrier than
the preceding step. Therefore the overall barrier for the
transformation of 1 into 5 is set by these C-H activation
processes. The hydride ligand of 1e then migrates onto the
Ccarbene atom (former Ccarbyne) leading to the 16 e- species 1f.
A subsequent C-H oxidative addition results in the new carbene
hydrido complex 1g. Reductive C-H elimination closes the
rearrangement to the TMP isomeric ring system to yield the
carbene complex 5b related to 1b. As for 1b and 1a, 5b is
expected to be in equilibrium with 5a, the cis-isomer of 5, from
which 5 is eventually generated by ligand rearrangement.
Various Inter- and Intramolecular Dynamics of 1 in the
Liquid and Solid State. HW/HW Exchange at Room Tem-
perature in Solution Traced by a Deuterium Double Label-
ing Crossover Study Converting 2 and 3 into 1 and 4. In the
previous paragraph the rearrangement of 1 into 5 was established
consisting of a sequence of different elementary process, which
are distinguished by their activation barriers. The room-
temperature steps lead to carbene formation from the carbyne
unit and the hydride ligand. The resulting 16e- complex could
indeed undergo a reversible “self-trapping” via dimerization with
concomitant exchange of the carbene units.
Figure 1. ORTEP plot of 5 with thermal ellipsoid probabilities of 50%.
All H atoms except for H49 are omitted for clarity.
Table 1. Selected Bond Lengths (Å) and Bond Angles (deg) for 5
Bond lengths
W(1)-C(13)
W(1)-P(3)
W(1)-P(2)
1.838(9) W(1)-P(4)
2.412(3) W(1)-P(1)
2.420(3) W(1)-H(49)
2.429(3)
2.436(3)
1.62
Bond angles
C(13)-W(1)-H(49) 168.1(3)
C(13)-W(1)-P(3)
P(3)-W(1)-P(2)
P(3)-W(1)-P(4)
93.1(3)
96.97(11)
81.19(11)
102.4(3)
80.88(10)
82.09(8)
91.85(7)
2
Double H-labeled crossover experiments were carried out
C(13)-W(1)-P(2)
C(13)-W(1)-P(4)
P(2)-W(1)-P(4)
P(3)-W(1)-P(1)
P(4)-W(1)-P(1)
P(2)-W(1)-H(49)
P(1)-W(1)-H(49)
98.5(3)
98.1(3)
in toluene at room-temperature variable but approximately equal
concentrations of the isotopomers 2 (W(CMes)(dmpe)2D) and
3 (W(CMes)(dmpe-d12)2H). The reaction was monitored by high
resolution 1H and 31P NMR spectroscopy allowing us to
distinguish all the potentially appearing isotopomers 1-4. The
observed scrambling indeed proceeded with equilibration into
all possible isotopomers (Scheme 3). It should be noted that
isotopomers possessing mixed labels in the phosphine ligands
were not observed so that a phosphine exchange process can
be ruled out.
163.37(10) C(13)-W(1)-P(1)
164.55(9) P(2)-W(1)-P(1)
96.47(10) P(3)-W(1)-H(49)
71.54(8)
82.74(7)
P(4)-W(1)-H(49)
C(14)-C(13)-W(1) 174.7(7)
toward the unsubstituted side. In 1 the mentioned repulsive
forces are balanced leading to an axially symmetric WtCs
CAr arrangement. All these observations indicate that 5 is
thermodynamically considerably more stabilized than 1. How-
ever, a thermodynamic relaxation process is expected to be
topologically complex and must consist of various consecutive
elementary steps.
In the high resolution 31P NMR spectrum the four isotopomers
(1-4) could be distinguished by their different chemical shifts
due to significant deuterium isotope effects: 2, 25.34 ppm (t);
1, 25.33 ppm (s); 4, 23.85 ppm (t); 3, 23.84 ppm (s). Moreover,
Mechanism for the Rearrangement of 1 into 5. In the
following sections we show that the mechanism for the
rearrangement of 1 into 5 can be rationalized on the basis of an
R-shift of the hydride ligand and a sequence of methyl oxidative
additions/reductive eliminations at the electron rich tungsten
center. A mechanism is sketched in Scheme 2 including an
isotopic labeling scheme to be discussed later.
1
1 and 3 can be distinguished in the H{31P} NMR spectrum
(toluene-h8) by distinct hydride signals, which are separated by
50 ppb. The presence of 1 and 3 has also been confirmed by a
P,H-correlation spectrum with a hydride resonance of 3 in
toluene-h8 at δ -6.67 ppm.
A concentrated mixture of 2 and 3 (15.5 mM) reached
equilibrium at room temperature within 8 days, while a mixture
of lower concentration (6.5 mM of 2 and 3) took ca. 46 days
for equilibration at under otherwise the same conditions. When
a 2.0 mM solution of 2 and 3 was used, exchange became so
It was observed earlier that a complex related to 1, namely
trans-W(CMes)(H)(CO)([P(OMe)3)]3, is dynamic at room tem-
16
perature and equilibrates to cis-W(CMes)(H)(CO)([P(OMe)3)]3
and the isomeric coordinatively unsaturated carbene complex
W(CHMes)(CO)[P(OMe3)]3. Based on this knowledge it appears
to be very plausible to assume that such equilibria would also
exist for 1 at room temperature and hence lead in subsequent
steps to 1a and 1b. At 60 °C or even somewhat below that
temperature reversible oxidative additions of the ortho
1
slow that it was no longer detectable either by H or by 31P
NMR spectroscopy. All these observations point to an exchange
mechanism with a rate-determining bimolecular step.
The McKay equation18 was applied to the kinetic pursuit of
the crossover experiments of 2 and 3 (Figure 2) and an
(17) Spivak, G. J.; Caulton, K. G. Organometallics 1998, 17, 5260.
(18) Espenson, J. H. In Chemical Kinetics and Reaction Mechanisms; 2nd ed.;
McGraw-Hill Inc.: New York, 1995; pp 55-57.
(16) Bannwart, E.; Jacobsen, H.; Furno, F.; Berke, H. Organometallics 2000,
19, 3605.
9
J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007 7197

A R T I C L E S
Scheme 2
Zou et al.
approximate first-order reaction was established as expected for
a self-exchange process which is based on the integrations of
the signal assigned to 3 at different times. This allowed us to
calculate an exchange rate of Rex ) 9.2 × 10-6 mol L-1 h-1
that could be calculated from this relationship. The exchange
reaction of 2 and 3 was observed to be even faster at lower
temperatures, which again supports an associative reaction
mechanism.
9
7198 J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007

trans-W(Cmesityl)(dmpe)2H: Polar W−H Bond and H-Mobility
A R T I C L E S
Scheme 3
Scheme 4
As proposed in Scheme 4, 2 and 3 would be in equilibrium
with the isomeric cis-vacant carbene complex 2b or 1bd. Then
in a bimolecular step the dinuclear intermediate 6 is expected
to be formed via association of these carbene molecules. 1 and
4 would result from the splitting of 6 in a way opposite to which
it was formed.
exchanges, which does however stress the existence of carbene
intermediates formed by a hydride to carbyne ligand shift
(Scheme 2).
Exchange of 1 with D2 Gas in Solution at Room Temper-
ature. The occurrence of a carbene intermediate may also be
made plausible in an H/D exchange experiment of 1 with D2
gas. The D2 molecule traps the carbene complex to form a D2
complex, which in turn is expected to undergo oxidative addition
and subsequent scrambling of metal bound deuterium into the
carbene hydrogens via reversible R-migration as sketched in
Scheme 2.
The observed “hydride” H/D exchange is thus not a direct
H/D exchange but rather a mediated process of carbene
The exchange experiments of 1 (W(CMes)(dmpe)2H) with
D2 gas were carried out in solution at various conditions of
pressure and temperature. When a solution of 1 was left at room
temperature under a pressure of 2 bar of D2, no apparent H/D
exchange according to Scheme 5 was detected even after 10
days. An increase in the D2 pressure to 60 bar caused the H/D
exchange to become noticeable after 6 days, but the amount of
2
2 formed was minor as verified by 31P and H NMR. Now a
Scheme 5
Figure 2. McKay plot for the exchange reaction of 2 and 3 to give 1 and
4 at room temperature (starting concentrations: 2, 6.5 mM; 3, 6.3 mM).
1H{31P} NMR pursuit of the decay of 2 with the concentration Ct reached
at equilibrium state with Ce ) 0.486 and F ) Ct/Ce (R2 ) 0.93).
9
J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007 7199

A R T I C L E S
Zou et al.
Figure 3. 31P NMR spectrum (toluene-h8, rt) of the reaction mixture obtained with heating 2 at 60 °C.
Scheme 6
very weak multiplet appearing at -6.62 ppm was then observed
in the H NMR spectrum.
ppm and -6.62 ppm, which were assigned to 5d and 1md,
2
respectively. The HP-CH resonances of 1md and 5d are close
3
to each other at 1.58 and 1.38 ppm.
In the corresponding 2H NMR spectrum two quintets appear
When a solution of 1 was pressurized with 105 bar of D2
gas at room temperature and left for 7 days, complete H/D
exchange occurred with full conversion into 2 (Scheme 5).
While the corresponding hydride resonance for 1 could no longer
be found in the 1H NMR spectrum, the 2H NMR spectrum, on
the other hand, showed an intense resonance for 2 at -6.57
ppm.
at -5.83 and -6.57 ppm, which are assigned to 7 and 2,
2
respectively. The H nucleus of 1md in the deuterated methyl
group of the mesityl moiety appears at 2.50 ppm as a triplet as
a result of coupling with the two methylene protons. Likewise,
the 2H nucleus of the deuterated methyl group of the TMP ring
in 5d is also assigned a triplet signal at 2.32 ppm because of
It was then of interest to see if the rearrangement of 1 into 5,
for which the 16e- carbene intermediate 1b was also postulated,
could be intersected by the reaction with D2 and equilibrated
to the isotopomers of 1 and 5. Indeed, when a Young tap NMR
tube filled with a hexane solution of 1 was pressurized with 2
bar of D2 gas and the reaction was run at 40 °C in hexane for
10 days, partial H/D exchange occurred to form the deuterides
5d (W[C(TMP)](dmpe)2D) and 2. Both species were detected
2
the same type of coupling. No H NMR signal is observed in
the possible shift range for P-CDH2 groups of 1.65 to 1.30
ppm, the possible shift range for P-CDH2 groups, which would
have indicated the presence of H/D exchange into the dmpe
ligands.
The intramolecular H/D exchange of 2 is also recognized by
31P NMR spectroscopy on account of differences based on small
but significant isotope shifts (Figure 3). A triplet at 23.83 ppm
is attributed to 7, while 5d gives rise to a singlet at 23.86 ppm;
another triplet at 23.30 ppm is assigned to 2, while the resonance
of 1md appears at 22.29 ppm. Additionally, two additional
singlets appear at 23.83 ppm (5) and 23.30 ppm (1), which are
partially overlapping with signals of 7 and 2, respectively. The
formation of 5 and 1 originates from the earlier described room-
temperature processes of intermolecular HW/DW exchanges of
7 and 2 with any hydride source. These H/D exchange processes
would be fully consistent with the pathway shown in Scheme
2 and are expected to occur at the stage of 1f/2f as the crucial
intermediate(s).
2
by H NMR spectroscopy. A resonance at -5.54 ppm was
assigned to 5d in addition to a quintet at -6.50 ppm belonging
to 2.
Intra- and Intermolecular H/D Exchanges of 1 in Solution
at 60 °C. According to Scheme 2 the hydride ligand of 1 would
be expected to show at elevated temperatures exchange with
the hydrogen atoms of the ortho-methyl groups of 1 before it
is eventually converted further into 5.
Indeed, when a solution of 2 was kept in toluene at 60 °C
for 3 days, a scrambling of the deuterium content into the ortho-
methyl groups of the mesityl moiety atom was observed
(Scheme 6).
The formation of the products 1md, 5d, and 7 given in
Scheme 6 was confirmed by NMR spectroscopy. In the 1H NMR
spectrum of the isotopic mixture two quintets appear at -5.88
Solid-State H-Mobility: Intramolecular H/H or H/D
Exchange of 1 or 2 at Elevated Temperatures. The H-mobility
in solution described in the previous chapters was in all cases
9
7200 J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007

trans-W(Cmesityl)(dmpe)2H: Polar W−H Bond and H-Mobility
A R T I C L E S
Scheme 7
induced by an initial rearrangement of the ligand sphere. Related
processes could not be anticipated to occur in the solid state
with reasonable activation barriers. If 1 were to exhibit any
isomerization with concomitant H-mobility, one would expect
this to take place at higher temperatures (>60 °C) and differ
from the processes described in the previous chapters.
Quasielastic Neutron Scattering.19 It was therefore of consider-
able interest to test if the solid state intramolecular H-mobility
described in the previous chapter could be studied in more detail
by this method. These measurements would not only give
evidence of the presence of such a H/H exchange process, but
these data would also provide information on the “jump length”
of the mobile H as well as an activation energy for this process
when the measurements are carried out as a function of
temperature. This seemed possible in a relatively large temper-
ature range, since 1 melts at 433 K.
The technique of Quasielastic Neutron Scattering (QNS) will
first of all provide evidence of the H-motion of 1 only if it is
sufficiently fast to result in a broadening of the elastic line given
by the available neutron scattering spectrometers. This method
has been used for a wide variety of related problems such as
proton conductors20 and molecular transport in layer com-
pounds22 and porous materials.23 QNS has, however, rarely been
used on organometallic compounds except for the observation
of reorientational motions of methyl or cyclopentadienyl
groups23 when these are the only H-containing ligands on the
metal. The only example of translational proton motion in an
organometallic compound observed24 by QNS has been, to the
best of our knowledge, that of H in the exchange of hydride
and dihydrogen ligands in the Ir complex IrClH2(η2-H2)(P-i-
Pr3)2.
Quasielastic Neutron Scattering data on approximately 2 g
of 1 were collected using the High Flux Backscattering
Spectrometer at the NIST Center for Neutron Research up to
temperatures of 425 K. The energy resolution of the instrument
was 1.3 µeV which corresponds to a frequency of about 3 ×
108 s-1. Broadening of the quasielastic line was only apparent
at temperatures above 300 K. At 150 K we observed quasielastic
broadening from the reorientations of the methyl groups, which
moved outside the very narrow energy window of this instru-
ment at higher temperatures and, hence, did not affect our
results. Quasielastic spectra (summed over all angles) are shown
for 25, 400, and 425 K in Figure 4. The angular, or Q
Indeed, when 5 mg of solid isotopmer 2 was placed into a
Young tap NMR tube and then heated to 100 °C for 2.5 h, the
room temperature 31P NMR spectrum in toluene-h8 gave
evidence for the formation of a mesitylidyne hydrido tungsten
1
species. The H NMR spectrum in toluene-h8 showed that a
complex similar to 1 had formed as indicated by a hydride
multiplet at -6.61 ppm. A typical signal was observed at 1.32
2
ppm in the H NMR spectrum which was assigned to mono-
deuterated P-CH2D methyl groups. An H/D exchange process
according to Scheme 7 was evidently taking place. The solid-
state structure of 1 reveals that the shortest distance of HW to
Hmethyl of the dmpe ligand is approximately 4.25 Å, which
indeed is very close to the “jump length” of the Quasielastic
Neutron Scattering experiment of 1 (vide infra) are in accord
with such an exchange process.
Since the atomic mobility in the solid state is restricted and
only hydrogen can normally show dynamic behavior, we have
to assume that the mechanism for the rearrangement of 2 into
1Pd could for instance not be envisaged to take place via
oxidative addition of an Mep group to the tungsten center,
because formation of a vacant site would be required. A vacant-
site species like 1b of Scheme 2 cannot be accessed due to a
too large degree of necessary topological changes. The pathway
of a rearrangement of the dmpe framework possesses however
a simple alternative utilizing an intramolecular deprotonation
process of 2 with the deuteride ligand as a base. This ligand is
expected to show quite a high “gas phase” basicity so that a
dmpe methyl group can be deprotonated, perhaps with the help
of an appropriate vibrational mode, to form an intermediate
dihydrogen complex. This dihydrogen intermediate represents
the crucial species for the H/D exchange, and “readdition” of
(19) Karmonik, C.; Hempelmann, R.; Matzke, T.; Springer, T. Z. Naturforsch.,
A: Phys. Sci. 1995, 50, 539-548.
-
D+ to the -CH2 local anion could complete this exchange.
(20) Poinsignon, C. Solid State Ionics 1997, 97, 399.
The same rearrangement is anticipated to be the basis of the
H-mobility detected in the Quasielastic Neutron Scattering
experiment of 1 described below.
(21) For example, see: Swenson, J. R.; Bergman, W. W.; Howells, S. J. Chem.
Phys. 2000, 113, 2873.
(22) Jobic, H. J. Phys. IV France 2000, 10, 125.
(23) Bee, M. Quasielastic Neutron Scattering; Adam Hilger: Bristol, U.K., 1988.
(24) Raison, P.; Lander, G. H.; Delapalme, A.; Williams, J. H.; Kahn, R.; Carlile,
C. J.; Kanellakopolum, B. Mol. Phys. 1994, 81, 309. Prager, M.; Grimm,
H.; Desmedt, A.; Lechner, R. E. Chem. Phys. 2003, 292, 161.
H-Mobility Studied by Quasielastic Neutron Scattering.
Hydrogen mobility can in principle be readily observed by
9
J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007 7201

A R T I C L E S
Zou et al.
Figure 4. Quasielastic spectra, summed over all angles, for trans-W(CMes)(dmpe)2H (1) at 25 (s), 400 (O), and 425 K (b).
dependence, of the quasielastic width is shown for the two
highest temperatures in Figure 5 along with fits of the two
models described below.
The average line width for the three temperatures (where the
point at T ) 25 K is taken to be resolution-limited) can be fit
to an Arrhenius law, with the result that the activation energy
for this process is about 5 kcal/mol.
We have attempted to fit two common models for jump
diffusion25 to these data, the Chudley-Elliott model with a
single jump distance as is commonly used in solids, and one
based on a distribution of jump lengths as one would encounter
in a liquid. Both models gave similar results with jump lengths
between 4.5 and 6.5 Å (Table 2). The residence time τ decreases
and the jump length increases from 400 to 425 K.
The single-crystal X-ray diffraction structure shows that the
distance of Hhydride to Ho-methyl of the mesityl group of 1 is 4.25
Å, which obviously matches very well with model 2 of the
Quasielastic Neutron Scattering experiments of 1. The alternative
direct “jump” of HW to CCarbyne to form a carbene complex
would amount to only about 3.7 Å. As a “trans jump” it would
however have to overcome high kinetic barriers and would not
be productive forming new isotopomers from 2. The generation
of 1Pd from 2 makes it necessary to assume the same process
operating in the Quasi Elastic Neutron Scattering Experiment.
The low activation barrier found for the solid state rearrangement
(5 kcal/mol) is somewhat in contrast to the general experiences
that related chemical processes have normally lower activation
energies in solution than in solid state. However, for intramo-
lecular H-mobilities this does not always seem to hold.
Finally it should be noted that an intermolecular HW/HW self-
exchange in the solid state is not supported by the jump length
determined from the quasielastic neutron scattering experiment,
since the HW‚‚‚HW nonbonding distance is 8.5 Å based on the
single-crystal X-ray diffraction study, as well as on a recent
single-crystal neutron diffraction study.26
Conclusions
A plausible mechanism could be established for the isomer-
ization of 1 into 5 using a variety of specific H/D exchange
and double crossover experiments. The reaction is expected to
be exothermic due to a release of steric strain. The kinetics of
this rearrangement can be separated into room-temperature
processes of a ligand sphere rearrangement and a hydride to
CCarbyne migration and a series of C-H oxidative addition/
reductive elimination processes of the ortho mesityl methyl
groups, the latter running at elevated temperature (60 °C).
Thermal treatment of 1 at 100 °C in the solid state causes H/H
exchange of the hydride ligand with the Medmpe groups being
thus in chemoselectivity different from those processes observed
in solution. A Quasielastic Neutron Scattering study of 1 in the
solid state confirmed that a H jump exists in the molecule with
a distance of 4-5 Å, which approximately matches the HW to
Medmpe distance (4.25 Å). The rearrangements and intermo-
lecular, as well as intramolecular, exchanges in liquid and in
the solid state thus primarily reflect the dynamics of the HW
atom of 1.
Experimental Section
General. All operations with air-sensitive complexes were performed
under an inert atmosphere of N2 using Schlenk and vacuum line
techniques or in a glovebox (Model MB-150B-G). All solvents
(including deuterated solvents) were dried and deoxygenated employing
appropriate drying/deoxygenating agents and then were newly distillated
under nitrogen before use (tetrahydrofuran, toluene, benzene, hexane,
pentane, ether, sodium/benzophenone; CH2Cl2, P2O5, and then CaH2).
IR spectra were obtained on a Bio-Rad FTS instrument. Raman spectra
were recorded on a Renishaw Ramanscope spectrometer (514 nm).
NMR spectra were recorded on the following spectrometers: Varian
(25) Li, S.; Hall, M. B.; Eckert, J.; Jensen, C. M.; Albinati, A. J. Am. Chem.
Soc. 2000, 122, 2903.
(26) Zou, F.; Albinati, A.; Berke, H. Unpublished results.
9
7202 J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007

trans-W(Cmesityl)(dmpe)2H: Polar W−H Bond and H-Mobility
A R T I C L E S
Figure 5. Q-dependence of the quasielastic line widths for trans-W(CMes)(dmpe)2H (1) at T ) 400 K (top) and T ) 425 K (bottom). The fitted lines
correspond to the two models for translational diffusion (see Table 2).
Table 2. Results for Fitting the Q-Dependence of the Quasielastic
Width to the Singwi-Sjolander (Model 1) and Chudley-Elliott
Forms (Model 2) for Translational Diffusion25
to that of dmpe using CD3I as reagent. W(CMes)(dmpe-d12)2H (3) was
obtained in a way analogous to that for W(CMes)(dmpe)2H using
dmpe-d12.
model 1
jump distance (Å)
model 2
jump distance (Å)
1. Preparation of W(CTMP)(dmpe)2H (TMP ) 2,3,5-trimeth-
ylphenyl) (5). A solution of 9.2 mg (0.0150 mmol) of W(CMes)-
(dmpe)2H (1) in 0.6 mL of toluene-d8 in a Young tap NMR tube was
heated at 60 °C for 2 weeks. 1H NMR and 31P NMR pursuit
demonstrated a 90% spectroscopic yield. After the solvent was removed
in Vacuo the remaining red solid was dissolved in pentane and filtered
through Celite. The pentane filtrate was cooled to -35 °C, upon which
red crystals of 5 (6.6 mg, 0.0107 mmol) were obtained (yield 71%).
Single crystals suitable for a single-crystal X-ray diffraction study
were obtained by crystallization from a pentane solution at -35 °C
(Table 3).
1H NMR (300 MHz, C6D6, 295 K): δ 7.02 (s, 1H, p-TMP), 6.73 (s,
1H, o-TMP), 2.41 (s, 3H, CH3sTMP), 2.24 (s, 3H, CH3sTMP), 2.10
(s, 3H, CH3sTMP), 1.71 (s, 12H, 4 × CH3sP), 1.51 (s, 12H, 4 ×
CH3sP), 1.55 (m, 8H, 4 × CH2sP, overlap with one of CH3sP group),
T (K)
τ
τ
400
425
5.6
6.4
0.15
0.14
4.5
4.6
0.2
0.17
Gemini-200 instrument (1H at 199.98 MHz, 13C at 50.29 MHz, 31P at
80.95 MHz), Varian Gemini-300 instrument (1H at 300.1 MHz,
2H at 76.7 MHz, 13C at 75.4 MHz, 31P at 121.5 MHz), Bruker DRX-
500 instrument (1H at 500.2 MHz, 13C at 125.8 MHz, and 31P at 202.5
MHz). δ (1H) and δ (13C) relative to SiMe4, δ (31P) relative to 85%
H3PO4, δ (11B) relative to Et2O‚BF3. The solid-state NMR measurements
have been carried out at -40 °C using a narrow bore probe for 4 mm
rotors.
The compounds W(CMes)(dmpe)2H(D)5a (1 and 2) were synthesized
according to the literature. dmpe-d12 was prepared in an analogous way
9
J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007 7203

A R T I C L E S
Zou et al.
Table 3. Summary of the X-ray Diffraction Study of 5
2.4. H/D Exchange Reaction in a Young Tap NMR Tube at 2
bar of D2. In a Young tap NMR tube a 5.4 mM solution of 1 in hexane
was pressurized with 2 bar of D2. The reaction mixture was left at rt
for 10 days. No exchange was observed based on the 31P and 2H NMR
spectra recorded.
empirical formula
formula weight (g‚mol-1
temperature (K)
wavelength (Å)
crystal syst, space group
a (Å)
C22H44P4W
616.30
183(2)
)
0.710 73
triclinic, P1h
10.0806(9)
12.6747(12)
12.8077(10)
73.376(10)
67.246(9)
66.989(10)
1371.3(2)
2, 1.493
4.451
620
0.23 × 0.05 × 0.02
2.28 to 25.88
10 799
2.5. H/D Exchange Reaction under 60 bar of D2 in an Autoclave.
3.3 mg of 1 were dissolved in 1 mL of toluene (5.4 mM) in an
autoclave, which was pressurized with 60 bar of D2 at 60 °C for 6
days. Traces of exchange were noticed, leading to the formation of 2.
2H NMR (76.6 MHz, toluene, rt): 2, δ -6.62 ppm (multiplet, DW).
2.6. H/D Exchange Reaction under 105 bar of D2 in an Autoclave.
16.6 mg of 1 was dissolved in 1 mL of toluene (27.0 mM) in an
autoclave, which was then pressurized with 105 bar of D2 gas. The
reaction mixture was stirred with a magnetic stirring bar at rt for 7
days. At the end of the reaction the remaining D2 gas was slowly
released under the protection of N2 gas. Then an NMR sample was
made of the reaction mixture for NMR measurements, which showed
a complete H/D exchange between HW of 1 and D2 gas affording 2. 2H
NMR (76.6 MHz, toluene, rt) δ -6.57 ppm (quintet, DW).
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
volume (Å3)
Z, density(calcd) (Mg/m3)
abs coefficient (mm-1
F(000)
)
crystal size (mm3)
collection θ range (deg)
reflections collected
reflections unique
θ(max) (deg), completeness
to θ (%)
absorption correction
max/min transmission
data/restraints/parameters
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
4983 [R(int) ) 0.0823]
25.88, 93.4
numerical
0.9027 and 0.7612
4983/0/241
3. H/D Exchange of 2 in Solution at 60 °C. In a Young tap NMR
tube 10.0 mg of 2 was dissolved in 0.5 mL of toluene (32.5 mM). The
sample was heated to 60 °C for 3 days. Then the exchange products
0.869
1
2
were identified by H, H, and 31P NMR spectroscopy.
1H NMR (500.0 MHz, toluene-h8, rt): 5d, δ -5.88 ppm (quintet,
HW), 1.38 ppm (broad doublet, HP-CH3); 1.1md, δ -6.62 ppm (quintet,
Hw), 1.58 ppm (broad doublet, HP-CH3).
R1 ) 0.0540, wR2 ) 0.1322a
R1 ) 0.0863, wR2 ) 0.1400b
a The unweighted R-factor R1 ) Σ(Fobsd - Fcalcd)/ΣFobsd b The weighted
.
R-factor wR2 ) Σw(F2
- F2calcd)2/Σw(F2obsd)2.
obsd
2H NMR (76.7 MHz, toluene-h8, rt): 7, -5.83 ppm (quintet, DW);
2, -6.57 ppm (quintet, DW); 1md, 2.50 ppm (t, HCH D of mesityl
2
2
1
-5.71 ppm (quintet, 1H, WsH, JPH ) 30 Hz, JWH ) 30 Hz). 31P
moiety), 5d, 2.32 ppm (t, HCH D of 2,3,5-trimethylphenyl moiety).
2
1
NMR (121.48 MHz, C6D6, 295 K): δ 25.2 ppm (s, JWP ) 75.8 Hz).
31P NMR (202.5 MHz, toluene-h8, rt): 7, 23.83 ppm (t); 5d, 23.86
ppm (s); 2, 23.30 ppm (t); 1md, 22.29 ppm (s); 5, 23.83 ppm (s, partially
overlap with 7); 1, 23.30 ppm (s, partially overlap with 2).
4. Intramolecular H/D Exchange in the Solid State at 100 °C. 5
mg of 2 (red powder) were put into a Young NMR tube, and then the
sample was heated at 100 °C for 2.5 h. After cooling to room
temperature the sample was dissolved in 0.5 mL of toluene-h8. A 31P
NMR spectrum showed that 1Pd was formed.
13C NMR (125.8 MHz, toluene-d8, 293 K): δ 262.0 (m, 1C, WtC),
155.5 (s, 1C, ipso-TMP), 129.2 (s, 1C, o-CHsTMP), 128.2 (s, 1C,
p-CHsTMP), 136.0 (s, 1C, o-CsTMP), 132.6 (s, 1C, m-CsTMP),
130.0 (s, 1C, m-CsTMP), 36.1 (m, 4C, 4 × CH2sP), 26.6 (m, 4C, 4
× CH3sP), 24.8 (m, 4C, 4 × CH3sP), 21.3 (s, 1C, CH3sTMP), 19.9
(s, 1C, CH3sTMP), 17.2 ppm (s, 1C, CH3sTMP).
Anal. Calcd for C22H44P4W: C, 42.87; H, 7.20. Observed: C, 43.20;
H, 7.26.
1H NMR (500.0 MHz, toluene-h8, rt): δ -6.61 ppm (m, very weak,
HW).
2. H/D Exchange Reactions. 2.1. Double Label Crossover Experi-
ment Probing HW/DW Exchange Using a Solution of 2 and 3 (15.5
mM) at Room Temperature. 0.3 mL of solution of 2 (31.0 mM) in
toluene was mixed with 0.3 mL of a solution of 3 with the same
concentration in toluene. The HW/DW exchange was observed shortly
after the solutions of 2 and 3 were mixed. The reaction reached
equilibrium within 8 days affording 1 and 4.
31P{1H}NMR (202.5 MHz, toluene-h8, rt): 2, δ 25.34 ppm (t); 1, δ
25.33 ppm (s); 4, δ 23.85 ppm (bt); 3, δ 23.84 ppm (bs).
1H{31P} NMR (500.0 MHz, toluene-h8, rt): 1, δ -6.61 ppm (s, HW);
3, δ -6.66 ppm (s, HW).
2H NMR (76.7 MHz, toluene-h8, rt): δ 1.32 ppm (DCH D of 1Pd),
2
-6.61 ppm (quintet, DW of 2).
31P NMR (202.5 MHz, toluene-h8, rt): 2, δ 25.32 ppm (t); 1Pd, δ
25.31 ppm (s).
5. X-ray Diffraction Study on 5. Crystals of 5 were grown under
an atmosphere of dry nitrogen in a glovebox and embedded in
polybutene oil. The only suitable crystal out of a manifold of very tiny
thin needles was selected for the X-ray experiment by using a polarizing
microscope and was mounted on the tip of a glass fiber and immediately
transferred to the goniometer of an imaging-plate-detector system (Stoe
IPDS diffractometer). The crystal was cooled to 183(2) K by using an
Oxford Cryogenic system. In order to suppress a dominating number
of unobserved intensity reflections, the crystal-to-image distance was
set to 70 mm (θmax ) 25.88°) and the exposure time was set to 6.0
min per image, resulting in a φ-oscillation scan mode for the data
collection. For the final cell parameter refinement, 8000 reflections were
selected from the whole limiting sphere. A total of 10 799 diffraction
intensities were collected,27 of which 4983 were unique (Rint ) 0.0823)
after data reduction. 3247 of these reflections were considered observed
by criterion I > 2σ(I). A numerical absorption correction28 based on
eight crystal faces was applied with FACEitVIDEO and XRED.27 The
structure was solved by direct methods using the program SHELXS-
2.2. Double Label Crossover Experiment Probing HW/DW Ex-
change Using a Solution of 2 and 3 (6.5 mM) at Room Temperature.
The intermolecular HW/DW exchange between a solution of 2 (6.5 mM)
and 3 (6.3 mM) arrived at equilibrium within ca. 46 days giving 1 and
4. For 31P and H NMR data, see section 2.1.
1
2.3. Double Label Crossover Experiment Probing HW/DW Ex-
change at -45 °C. A 0.016 M solution of 2 and a 0.022 M solution
of 3 in toluene-d8 were independently cooled to -70 °C. Then 0.3 mL
of each solution of 2 and 3 was mixed in a Young tap NMR tube at
-70 °C when NMR spectra were then recorded at various temperatures.
Shortly after the mixing at -70 °C (∼ 5 min) intermolecular
exchange already started. After 35 min at -40 °C the reaction had
nearly arrived at equilibrium.
1H{31P} NMR (500.0 MHz, toluene-h8, -40 °C and -70 °C): 1, δ
-6.67 ppm (s, HW); 3, δ -6.73 ppm (s, HW). For the 31P NMR data of
the four isotopomers, see section 2.1.
(27) Stoe IPDS software for data collection, cell refinement, and data reduc-
tion: Version 2.92 Stoe, Darmstadt, Germany, 1999.
(28) Coppens, P.; Leiserowitz, L.; Rabinovich, D. Acta Crystallogr. 1965, 18,
1035.
9
7204 J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007

trans-W(Cmesityl)(dmpe)2H: Polar W−H Bond and H-Mobility
A R T I C L E S
97.29 Interpretation of the difference electron density maps, preliminary
plot generations, and checking for higher symmetry were performed
by using PLATON.30 The full-matrix least-squares refinements (based
on F2) were carried out with SHELXL-9731 using anisotropic displace-
ment parameters for all non-hydrogen atoms, except for atoms C13,
C16, and C17 which had to be refined isotropically due to nonpositive-
definite anisotropic displacement parameters. This observation can be
explained by many extreme tiny intergrown needles that contributed
and probably overlapped some intensity data, adding systematic error
to the absorption correction. The hydride H-atom position was
interpreted from difference maps and its coordinates were fixed, but
the isotropic displacement parameter was allowed to be refined.
Positions of all other hydrogen atoms were calculated after each
refinement cycle. The plot of the structure was generated by using
ORTEP.32
Acknowledgment. Financial support for this work from the
Swiss National Science Foundation and from the University of
Zu¨rich is gratefully acknowledged. Work at Los Alamos
National Laboratory was supported by the Office of Science,
U.S. Department of Energy.
Supporting Information Available: CCDC 627509 contains
the supplementary crystallographic data for this paper. These
data can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
quest/cif. 1H and 2H NMR spectra of the H/D exchange reaction
according to Scheme 6. This material is available free of charge
via Internet at http://pubs.acs.org.
JA069140K
(29) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467.
(30) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7-13.
(31) Sheldrick, G. M. SHELXL-97: Software Package for Crystal Structure
Determination and Refinement, University of Go¨ttingen, Go¨ttingen,
Germany, 1997.
(32) Johnson, C. K. ORTEPII, Report ORNL-5138, Oak Ridge, Tennessee,
U.S.A.
9
J. AM. CHEM. SOC. VOL. 129, NO. 22, 2007 7205