Hydridic reactivity of W(CO)(H)(NO)(PMe3)3 - Dihydrogen bonding and H2 formation with protic donors

Article abstract of DOI:10.1016/j.jorganchem.2009.10.036

The hydridic reactivity of the complex W(CO)(H)(NO)(PMe₃)₃ (1) was investigated applying a variety of protic donors. Formation of organyloxide complexes W(CO)(NO)(PMe₃)₃(OR) (R = C₆H₅ (2), 3,4,5-Me₃C₆H₂ (3), CF₃CH₂ (4), C₆H₅CH₂ (5), Me (6) and iPr (7)) and H₂ evolution was observed. The reactions of 1 accelerated with increasing acidity of the protic donor: Me₂CHOH (pK_a = 17) < MeOH (pK_a = 15.5) < C₆H₅CH₂OH (pK_a = 15) < CF₃CH₂OH (pK_a = 12.4) < C₆H₂Me₃OH (pK_a = 10.6) < C₆H₅OH (pK_a = 10). Regioselective hydrogen bonding of 1 was probed with two of the protic donors furnishing equilibrium formation of the dihydrogen bonded complexes ROH···HW(CO)(NO)(PMe₃)₃ (R = 3,4,5-Me₃C₆H_2, 3a and iPr, 7a) and the O_NO hydrogen bonded species ROH···ONW(CO)(H)(PMe₃)₃ (R = C₆H₂Me_3, 3b and iPr, 7b) which were studied in hexane and d₈-toluene solutions using variable temperature IR and NMR spectroscopy. Quantitative IR experiments at low temperatures using 3,4,5-trimethylphenol (TMP) confirmed the two types of competitive equilibria: dihydrogen bonding to give 3a (ΔH₁ = -5.8 ± 0.4 kcal/mol and ΔS₁ = -15.3 ± 1.4 e.u.) and hydrogen bonding to give 3b (ΔH₂ = -2.8 ± 0.1 kcal/mol and ΔS₂ = -5.8 ± 0.3 e.u.). Additional data for the hydrogen bonded complexes 3a,b and 7a,b were determined via NMR titrations in d₈-toluene from the equilibrium constants K_(Δδ) and K_{(Δ R1)} measuring either changes in the chemical shifts of H_W(_Δδ) or the excess relaxation rates of H_W (ΔR₁) (3a,b: ΔH_(Δδ) = -0.8 ± 0.1 kcal/mol; ΔS_(Δδ) = -1.4 ± 0.3 e.u. and Δ H_{(Δ R1)} = -5.8 ± 0.4 kcal/mol; Δ S_{(Δ R1)} = -22.9 ± 1.9 e.u) (7a,b: ΔH_(Δδ) = -2.3 ± 0.2 kcal/mol; ΔS_(Δδ) = -11.7 ± 0.9 e.u. and Δ H_{(Δ R1)} = -2.9 ± 0.2 kcal/mol; Δ S_{(Δ R1)} = -14.6 ± 1.0 e.u). Dihydrogen bonding distances of 1.9 A? and 2.1 A? were derived for 3a and 7a from the NMR excess relaxation rate measurements of H_W in d₈-toluene. An X-ray diffraction study was carried out on compound 2.

Full text of DOI:10.1016/j.jorganchem.2009.10.036

Journal of Organometallic Chemistry 695 (2010) 382–391
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Hydridic reactivity of W(CO)(H)(NO)(PMe3)3 – Dihydrogen bonding and H2
formation with protic donors
*
´
Nataša Avramovic, Jürgen Höck, Olivier Blacque, Thomas Fox, Helmut W. Schmalle, Heinz Berke
Department of Inorganic Chemistry, University of Zürich, Winterthurerstrasse 190, 8057 Zürich, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 July 2009
Received in revised form 19 October 2009
Accepted 25 October 2009
Available online 30 October 2009
The hydridic reactivity of the complex W(CO)(H)(NO)(PMe3)3 (1) was investigated applying a variety of
protic donors. Formation of organyloxide complexes W(CO)(NO)(PMe3)3(OR) (R = C6H5 (2), 3,4,5-Me3C6H2
(3), CF3CH2 (4), C6H5CH2 (5), Me (6) and iPr (7)) and H2 evolution was observed. The reactions of 1 accel-
erated with increasing acidity of the protic donor: Me2CHOH (pKa = 17)
< MeOH (pKa = 15.5) <
C6H5CH2OH (pKa = 15) < CF3CH2OH (pKa = 12.4) < C6H2Me3OH (pKa = 10.6) < C6H5OH (pKa = 10).
Regioselective hydrogen bonding of 1 was probed with two of the protic donors furnishing equilibrium
formation of the dihydrogen bonded complexes ROHꢀꢀꢀHW(CO)(NO)(PMe3)3 (R = 3,4,5-Me3C6H2, 3a and
iPr, 7a) and the ONO hydrogen bonded species ROHꢀꢀꢀONW(CO)(H)(PMe3)3 (R = C6H2Me3, 3b and iPr, 7b)
which were studied in hexane and d8-toluene solutions using variable temperature IR and NMR spectros-
copy. Quantitative IR experiments at low temperatures using 3,4,5-trimethylphenol (TMP) confirmed the
Dedicated to Prof. Ingo-Peter Lorenz on the
occasion of his 65th birthday. Department of
Chemistry and Biochemistry, Ludwig-
Maximillians-Universität München
Keywords:
two types of competitive equilibria: dihydrogen bonding to give 3a (
ꢁ15.3 1.4 e.u.) and hydrogen bonding to give 3b ( H2 = ꢁ2.8 0.1 kcal/mol and
Additional data for the hydrogen bonded complexes 3a,b and 7a,b were determined via NMR titrations
in d8-toluene from the equilibrium constants K( and K measuring either changes in the chemical
D
H1 = ꢁ5.8 0.4 kcal/mol and
DS1 =
S2 = ꢁ5.8 0.3 e.u.).
Hydrogen bonding
Dihydrogen bonding
Transition metal hydride
Tungsten
D
D
Dd)
ðDR1Þ
shifts of HW
(
d) or the excess relaxation rates of HW
(
D
D
R1) (3a,b:
D
H( d) = ꢁ0.8 0.1 kcal/mol;
D
S(
=
D
D
Dd)
Alkoxide
ꢁ1.4 0.3 e.u. and
D
H D = ꢁ5.8 0.4 kcal/mol;
S D = ꢁ22.9 1.9 e.u) (7a,b:
ð
D
H( d) = ꢁ2.3
ð
R1
Þ
R1
Þ
D
0.2 kcal/mol;
D
S( d) = ꢁ11.7 0.9 e.u. and
D
H D = ꢁ2.9 0.2 kcal/mol;
D
S D = ꢁ14.6 1.0 e.u). Dihy-
D
ð
R1
Þ
ð R1Þ
drogen bonding distances of 1.9 Å and 2.1 Å were derived for 3a and 7a from the NMR excess relaxation
rate measurements of HW in d8-toluene. An X-ray diffraction study was carried out on compound 2.
Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction
ature. Different methods might be applied to study dihydrogen
bonding of transition metal hydride complexes. The superior meth-
Proton transfers from protic donors to metal hydrides are
thought to be preceded by hydrogen bonding, in particular dihy-
drogen bonding between the hydride and the protic substrate
[1–31]. In conjunction with catalysis, such as ‘‘ionic hydrogena-
tions” proceeding with proton and hydride transfers [32–37], the
significance of particularly dihydrogen bonding is still a matter of
dispute. Intramolecular hydrogen bonding of the hydride ligand
was first discovered in the solid state in the groups of Morris and
Crabtree [9–11], while the groups of Shubina and Berke provided
spectroscopic, thermodynamic and structural evidence for inter-
molecular MHꢀꢀꢀHX bonding in solution [12–16].
odologies to examine these interactions include variable tempera-
ture (VT) IR and NMR spectroscopic studies, X-ray and neutron
diffraction studies and supporting density functional theory
(DFT) calculations. Crabtree’s group observed intermolecular dihy-
drogen bonding in the crystal of the rhenium complex Re-
H5(PPh3)3ꢀindole by an X-ray diffraction study [38–41]. Based on
HꢀꢀꢀH bond distances and their corresponding bond angles, the
strength of the HꢀꢀꢀH bond and linearity of the HꢀꢀꢀHRe fragments
could be established. In comparison to the solid state, VT IR spec-
troscopy carried out in solution allows observation of unique
hydrogen bonded species in equilibrium. Dihydrogen bonding
established with many tungsten and rhenium hydrides and acidic
alcohols like phenol, hexafluoro-2-propanol (HFIP) and perfluoro-
2-methyl-2-propanol (PFTB) was studied in our group using VT
IR spectroscopy [12,13,27]. Various other VT IR studies showed
that dihydrogen bonding species indeed precede the proton
transfer processes [42–44]. Examples are the protonations of [Re-
(CO)(NO)(PMe3)2H2] [15] and [CpRuH(CO)(PCy3)] [14,45,46] with
alcohols. In the latter case full protonation occurred with an
Tracing intermolecular dihydrogen bonding between metal hy-
drides [M]ꢁH and protic donors A–H is a difficult challenge, since
the involved attractive forces are similar in magnitude to the coun-
teracting entropy terms of the association process at room temper-
* Corresponding author. Fax: +41 1 63 56 802.
E-mail address: hberke@aci.uzh.ch (H. Berke).
0022-328X/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2009.10.036

´
383
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
intermediate identified as the ion pair [CpRu(CO)(PCy3)(g2
-
hol, 2,2,2-trifluoroethanol, methanol and isopropanol, respec-
tively. For the reactions with phenol and TMP two equivalents
were used, but in the cases of the generally less acidic aliphatic
alcohols, three equivalents of the alcohols were needed to drive
the reactions to completion within a reasonable span of time.
Plots of the decay of 1 vs. time t allowed an exponentional fitting
with a first order law. A mechanism could not be demonstrated
from these studies nevertheless, it is clear that these reactions
do not involve a rate determining bimolecular step (second order)
of the hydride with the alcohols. In agreement with the evidence
to be described later, we assume the involvement of dihydrogen
bonding pre-equilibria and subsequent rate determining first or-
der decays of the formed dihydrogen bonding adducts (Scheme 1).
The half-life times of these reactions to 2–7 were 46, 57, 216, 218,
275 and 815 min. A plot of these half-life times vs. pKa of the cor-
responding protic donor confirms gross dependence on the acid-
ity of the protic donor, i.e. predominant electronic influence
(Fig. 1). Steric influence was apparently prevailing in the case of
the reaction of 1 with isopropanol being somewhat outside of
the correlation window of the fairly linear dependence of the
other protic donors of Fig. 1. Isopropanol possesses a much
slower reaction rate.
H2)]+ꢀꢀꢀ[OR]ꢁ. Additional stabilization originated from the contact
between the dihydrogen ligand and the counter-ion. Structural
elucidation of quite a few examples of these intermediates allowed
for the generalization of the mechanism of the proton transfer pro-
cesses under consideration [47,48,3]. NMR spectroscopic evidence
for the MHꢀꢀꢀHX hydrogen bonding is acquired by upfield shifts of
the hydride resonance and a decrease of the relaxation times
(T1min) [12]. Since such equilibrations are usually fast on the
NMR time scale, respective resonances of the hydride species aver-
age and shift with the position of the equilibria. The equilibrium
chemical shift Dhydride, which is thus dependent on the concentra-
tions of the hydride and the protic donor HX and on temperature,
allows one to determine the equilibrium constants by curve fitting
of the NMR titration. Short HꢀꢀꢀH distances, less than the sum of the
van der Waals radii, were determined by NMR spectroscopy for
intermolecular dihydrogen bonding in solutions of W(CO)2(NO)(P-
Me3)2H and Re(CO)(NO)(PMe3)2H2 in the presence of the relative
strongly acidic alcohol components (CF3)2CHOH (HFIP) and
(CF3)3COH (PFTB) [13,49]. In addition to this, r(H H) distances
ꢀꢀꢀ
were also derived by measuring T1 relaxation times for rhenium
and ruthenium complexes [50,51].
In the last 15 years our group has reported the syntheses of a
series of transition metal hydrides [52–80]. One of the main goals
was to establish correlation between hydridicity of the hydride li-
gands and their activity in hydride transfers. As just derived the
hydridicity of hydride ligands is one of the major parameters
for dihydrogen bonding and the subsequent reaction of the protic
donor with the hydrides [77–80]. Acidity and alcohol concentra-
tions, temperature and the type of solvent played a major role
in these processes [50,53,80]. We anticipated that both types of
reactions might follow the same electronic principles. The tung-
sten hydride W(CO)(H)(NO)(PMe3)3 was studied in our group
for hydride transfer reactions. Apparently due to the presence of
The IR spectra of compounds 2–7 displayed
bands in the region of 1500–2000 cmꢁ1. While the
of these compounds showed a quite small spread of 5 cmꢁ1, the
(NO) bands appeared between 1584 and 1565 cmꢁ1and were
more sensitive to the nature of their trans organyloxide moieties
[64] showing increasing wavenumbers in the order of
7 < 6 < 5 < 4 < 3 < 2. An almost linear regression was found for the
m
(NO) and
m(CO)
m
(CO) bands
m
m(NO) bands with the pKa of the reacting protic donors (Fig. 2)
proving in this order lower basicity of the organyloxy residue.
The 1H NMR spectra supported the structures of the organylox-
ide complexes further. For all compounds characteristic resonances
could be attributed to the OCH moieties appearing as a quartet at
3.87 ppm for 4, as singlets at 4.86 for 5 and at 3.67 for 6 and as a
septet at 3.78 ppm for 7. In addition, the 13C{1H} NMR spectra of
4–7 displayed characteristic signals for the OCH carbon atoms at
71.8, 75.8, 62.3 and 70.9 ppm and signals for the CO ligands at
around 229 ppm. Crystals of 2 suitable for X-ray diffraction study
were grown from a saturated pentane solution at ꢁ30 °C. The
structure displays a tungsten center with pseudo-octahedral coor-
dination. The nitrosyl group is located trans to the alkoxide ligand
and the three PMe3 ligands lie together with the CO ligand in a
plane (Fig. 3).
three strong phosphine
r-donors and the nitrosyl group in trans
position to the hydride ligand, W(CO)(H)(NO)(PMe3)3 shows a
strong hydridic polarization of the M–H bond [64,78]. In support
of such hydridic hydride reactions, we wanted to generate more
quantitative data, which were acquired by aid of IR and VT
NMR spectroscopies and in suitable cases also X-ray diffraction
studies.
2. Results and discussion
2.1. Reaction of W(CO)(H)(NO)(PMe3)3 (1) with a variety of alcohols
It was mentioned before that the reactions of hydride com-
plexes with protic donors to yield organyloxide complexes and
H2 presumably proceed via the intermediacy of dihydrogen bond-
ing adducts as contact pairs (K1). The overall reaction rates may be
related to stereoelectronic influences of the protic donors via the
tightness of the K1 equilibria (adduct concentrations) and the bar-
riers of the subsequent reactions represented by k(H2). We thought
we could quantitatively probe such reactivity of 1 with phenols
and alcohols of different steric bulks and acidities, such as phenol
(pKa = 10), 3,4,5-trimethylphenol (pKa = 10.6), 2,2,2-trifluoroetha-
nol (pKa = 12.4), benzyl alcohol (pKa = 15), methanol (pKa = 15.5)
and isopropanol (pKa = 17) and look for correlation of the reactions
of 1 with the alcohols and their acidities. Using quantitative 1H and
31P1H NMR, such reactions were followed in toluene-d8 at 60 °C
yielding the organyloxide complexes (ON)(OC)(Me3P)3W(OR)
(R = C6H5 (2), 3,4,5-Me3C6H2 (3), CH2CF3 (4), CH2C6H5 (5), Me (6),
Me2CH (7)) and H2.
NO
W
NO
W
PMe3
PMe3
OC
PMe3
PMe3
OC
K1
+
ROH
toluene-d8,60°C
Me3P
Me3P
H
H
HOR
1
k(H2)
NO
W
PMe3
PMe3
OC
+ H2
Me3P
OR
R = C6H5 (2), 3,4,5-Me3C6H2 (3), CH2CF3 (4),
CH2C6H5 (5), Me (6), Me2CH (7)
The reactions were complete after approximately 4, 5, 20, 24,
24 and 72 h for phenol, 3,4,5-trimethylphenol (TMP), benzyl alco-
Scheme 1.

´
384
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
18
16
14
12
10
8
isopropanol
methanol
benzyl alcohol
2,2,2-trifluoroethanol
3-fold excess
2-fold excess
3,4,5-trimethylphenol
phenol
0
100
200
300
400
500
600
700
800
900
half-life time (min)
Fig. 1. Correlation plot between the half-life time of the tranformations of 1 with
the indicated alcohols to give 2–7 and the pKa of the alcohols.
Fig. 3. View of the molecular structure of compound 2 (ellipsoids drawn at the 30%
probability level). All hydrogen atoms have been omitted for clarity. Selected bond
lengths (Å) and bond angles (°): W1–N1 1.800(3), W1–C1 2.009(4), W1–O3
2.102(3), W1–P1 2.5113(9), W1–P2 2.4996(9), W1–P3 2.5502(11); W1–N1–O1
177.4(3), N1–W1–O3 171.83(13), W1–O3–C2 134.8(2).
1585
2
4
3
1580
1575
1570
1565
2.3. IR spectroscopic investigations
5
For the IR measurements in the low temperature regime, cold
solutions of 1 and the protic donors were mixed at 213 K and
transferred into a pre-cooled VT IR cell (d = 0.1 cm). The addition
of isopropanol or benzyl alcohol to 1 in hexane caused practically
no changes in the IR spectra. We concluded that hydrogen bonding
of any kind appeared only to a small extent and adduct formation
occurred in concentrations below the IR detection limits.
6
7
10
11
12
13
14
15
16
17
18
However, in the case of the mixture of 1 with 3.2 equivalents of
3,4,5-trimethylphenol (TMP) strong changes became visible in the
IR spectrum between 1500 and 2000 cmꢁ1 in the temperature
range of 213–293 K (Fig. 4). A reference spectrum of 1 was practi-
cally temperature independent showing three strong bands at
1895, 1618 and 1531 cmꢁ1. In the presence of TMP the adducts
3a and 3b were formed interacting with 1 at the hydride and the
ONO site (K1 and K2 of Scheme 2). The IR spectra showed overlaid
pKa
Fig. 2. Correlation between m
(NO) of the organyloxide complexes 2–7 (cmꢁ1) and
the pKa of their conjugate alcohols.
2.2. Hydrogen bonding of 1 with protic donors: VT IR and VT NMR
studies
m
(CO) bands of both species falling together in a band with in-
creased half-height width in comparison with the (CO) band of
1. The band maximum of the superimposed (CO) bands shifted
m
It was mentioned before that dihydrogen bonding adducts char-
acterized by K1, but also hydrogen bonding adducts to ONO (K2),
may precede the formation of the organyloxy products according
to Scheme 1. Therefore we found it intriguing to trace the nature
of these hydrogen bonding equilibria between 1 and the protic do-
nors by quantitative variable temperature IR and NMR methodolo-
gies [2–6,12–28]. The experiments were carried out in hexane
using protic donors of substantially different pKa values (isopropa-
nol, pKa = 17; benzylalcohol, pKa = 15 and 3,4,5-trimethylphenol,
pKa = 10.6) in the temperature range from 213 to 293 K. The short
time scale set by the IR pursuit was expected to allow detection of
the hydrogen bonded intermediates. Contrary to this, we expected
the equilibrations in the NMR experiments to be fast on the NMR
timescale with concomitant signal averaging of the involved
species.
m
to higher wavenumbers with decreasing temperature (maximum
shift of 9 cmꢁ1 with respect to the corresponding band of 1 at
213 K). This relatively small m(CO) shift suggests that the closer
environment of the CO group changes only marginally upon TMP
addition excluding that the OCO atom gets involved in hydrogen
bonding. Hydrogen-bonding interaction at the OCO site would any-
way be expected to cause a band shift in the direction opposite to
what was observed [29]. Similarly when the tungsten center of 1
would get involved in hydrogen bonding with a proton donor,
the
m
(CO),
m
(WH) and
m
(NO) bands are expected to be shifted to-
ward higher wave numbers by
D
m
= +100–150 cmꢁ1 [18–
20,28,29]. No such bands were observed in the IR spectra in the
presence of TMP concluding that such hydrogen bonding is not
occurring.

´
385
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
Fig. 4. IR spectrum of 1 at 213 K in hexane (black) and IR spectra of 1 in the presence of TMP (0.0173/0.0567 mol/L) at 213 (green), 233 (blue) and 293 K (red) in hexane.
Wavenumbers of the major bands are given in brackets. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)
PMe3
W
PMe3
W
PMe3
W
PMe3
H
PMe3
PMe3
H
K2
K1
ROH
+ ROH
ON
OC
ON
OC
H
ON
OC
HOR
PMe3
PMe3
PMe3
3a, 7a
1
3b, 7b
OH
(3), Me2CHOH (7)
Me
ROH =
Me
Me
TMP
Scheme 2.
The adduct 3b with hydrogen bonded TMP attached to the NO
with hydrogen bonding to the ONO atom (Fig. 5) [29]. The intensity
group causes indeed a new
m
(NO) band at considerably lower
of the band of 3a increases at the expense of the intensity of the
wavenumbers appearing for all temperatures at 1502 cmꢁ1
(Fig. 6). This band of 3b is of low intensity and is located shoul-
der-like at the flank of a neighbouring strong band of lower wave-
m
(WH) band of 1 upon lowering of the temperature. These intensity
changes are reversible indicating that the adduct formation is an
equilibrium. The (WH) band of 3b appearing as a shoulder to
the (WH) of 1 persists at 293 K. Its temperature dependence
was somewhat unexpected, since upon lowering of the tempera-
ture it decreases in intensity, while the (WH) band of 3a gets
m
numbers. The strong shift of the band of 3b with respect to
of 1 (–29 cmꢁ1) speaks for hydrogen bonding to the ONO atom [29].
The VT IR spectra of the (NO) region provide vague additional evi-
m
(NO)
m
m
m
dence for the dihydrogen binding equilibrium of TMP with 1 to
form 3a. Since hydrogen bonding to HW can not cause much elec-
stronger and prevails at low temperature at the expense of the cor-
responding band of 3b. This can be explained in terms of a compe-
tition of the 1/3a and 1/3b equilibria competing for 1. The
association process to form 3a shows the stronger temperature
dependence due to a larger entropic term (vide infra). Apparently
3a possesses a stronger dihydrogen bond and concomitant with
this it achieves also a tighter solvate packing.
tronic perturbation at the NO ligand, the
pected to appear in relative close vicinity to the
of 1. In the 293 K IR spectrum, for instance, a (NO) band appears
at 1542 cmꢁ1, which is tentatively assigned to 3a. Some ambiguity
in this assignment arises from the fact that the (NO) bands of
m
(NO) band of 3a is ex-
m(NO) absorption
m
m
compounds of type 1 have generally low intensities and that spe-
cies like 1, 3a and 3b appear in lowered equilibrium concentrations
so that the observed bands are at the detection limits.
The 1/3a and 1/3b equilibria according to Scheme 2 were then
quantitatively analyzed between 253 and 293 K via deconvolution
of the overlapping m(WH) bands and subsequent integration. Cal-
Analyzing the
m
(WH) region of the VT IR spectra, significant
culational details of the 1/TMP system for the determination of
the equilibrium constants K1 of the 1/3a and K2 of the 1/3b equilib-
ria in hexane can be found in the Supplementary material.
K1 and K2 belonging to the two competitive processes of Scheme
2 were analyzed as a function of temperature using Eqs. (1) and (2).
changes were seen upon TMP addition. Three overlapping bands
were detected attributed to the band of free 1 at 1618 cmꢁ1, to a
band for the W–HꢀꢀꢀHOR dihydrogen bonded adduct 3a at
1599 cmꢁ1 and a shoulder at around 1625 cmꢁ1 attributed to 3b

´
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
386
2.4. 1H NMR spectroscopic investigations
1H NMR spectroscopy is well-suited for probing hydrogen
bonding through hydride ligands using titrations as quantitative
experiments [9–11,23,27,29]. The equilibration processes of
Scheme 2 were expected to be fast on the NMR time scale with
concomitant signal averaging. We studied the cases of the interac-
tion of 1 with isopropanol and TMP as weak and strong protic do-
nor binding cases.
First we pursued changes in the chemical shifts by 1H NMR
spectroscopy occurring upon addition of the protic donors to 1.
Solutions of 1 were pre-mixed with various ratios of isopropanol
and TMP at temperatures below ꢁ55 °C and then the chemical
shifts of the hydride resonance were determined at various tem-
peratures. The plots of the chemical shift differences (Dd) vs. T(K)
are presented in the Supplementary material (Figs. S2 and S3).
The averaged signals of the hydrogen bonded adducts 7a,b with
isopropanol in toluene-d8 showed upfield shifted
became more pronounced when the concentration of isopropanol
was increased or the temperature lowered. The d values of 3a,b
displayed higher values of d in comparison with 7a,b apparently
Dd values, which
Fig. 5. Enlargement of the IR spectra in the m(WH) region of 1 (0.0173 mol/L) at
213 K in hexane (black) and of 1 in the presence of 3.2 eq of TMP (0.0567 mol/L) at
213 K (green), 233 K (blue), 243 K (purple) and 293 K (red). (For interpretation of
the references to colour in this figure legend, the reader is referred to the web
version of this article.)
D
D
due to a stronger interaction with the hydride ligand.
T1 1H NMR experiments of 1 with isopropanol and TMP exhib-
ited substantially shorter T1 of the HW nucleus, once the acidic sub-
strate was added (see Figs. S4 and S5 in Supplementary material),
with higher concentrations of the alcohols, the minima of the pseu-
do-parabolae shift toward higher temperatures and shorter relax-
ation times. The T1 values were significantly shorter for the
addition of TMP. This reflects a larger equilibrium shift according
to Scheme 2 toward the dihydrogen bonded adducts with TMP
than with isopropanol. These T1 results obtained in the presence
of different concentrations of the alcohols are fully consistent with
the conclusions reached from the chemical shift experiments.
VT NMR spectroscopy can also provide thermodynamic data
and geometrical parameters of dihydrogen bonding [27,29] at-
tained via titrations, in our case with changes of the concentrations
of the added protic donors. Changes in the equilibrium positions of
K1 ¼ ½3aꢂ=½1ꢂ½ROHꢂ ¼ expðꢁ
D
H01=RT þ
D
S01=RÞ
ð1Þ
ð2Þ
K2 ¼ ½3bꢂ=½1ꢂ½ROHꢂ ¼ expðꢁ
D
H02=RT þ
D
S02=RÞ
At all temperatures measured K1 is larger than K2 indicating pre-
dominance for K1 and hydrogen bonding between HW of 1 and
TMP (dihydrogen bonding) (Fig. 6). At 253 K for instance, K1 is ap-
prox. 3 times larger than K2. The temperature dependencies of
ln(K1) and ln(K2) vs. 1/T according to Eqs. (1) and (2) allowed to ex-
tract enthalpies:
D
H1 = ꢁ5.8 0.4 and
D
H2 = ꢁ2.8 0.1 kcal/mol
and entropies:
D
S1 = ꢁ15.3 1.4 and
D
S2 = ꢁ5.8 0.3 e.u. The ꢁ
DS
values lie in the typical range of 5–20 e.u. reported for hydrogen
bonding in organic systems [2,12,29,81,82] and additionally stress
that dihydrogen bonding in 3a is prevalent over hydrogen bonding
in 3b and has the higher temperature gradient.
Scheme 2 would be expressed as chemical shift changes (Dd meth-
od), while in the case of the 1H T1 NMR measurements changes in
the equilibrium positions would be expected to cause changes in
the excess relaxation rates
(D
R1(min) = 1/T1(min)(add)
–
1/
T1(min)(WH)) ( R1(min)). Equilibrium constants K(
D
and K
D
d)
ðDR1Þ
were obtained via pursuit of the changes in the chemical shifts of
the HW signal ( d) or of the excess relaxation rates ( R1(min))
and via curve fitting of the plots of these parameters of HW vs. con-
3.8
3.6
3.4
3.2
3.0
2.8
2.6
2.4
2.2
2.0
1.8
D
D
centrations of the protic donors. All hydride components of the
equilibria of Scheme 2 average to one HW signal. The
is expected to show response for both equilibria of Scheme 2, while
the R1(min) method was expected to be sensitive merely to
Dd method
D
changes of the closer chemical environment of the hydride ligand
of 1, thus specifically addressing interactions with HW of 3a and
7a only, while the equilibrium interactions with ONO of 7b and
3b are expected to stay with this method ‘‘silent”. The strength
of hydrogen bonding is solvent dependent and in more polar sol-
vents the equilibria are expected to shift toward the dissociated
side. Since relatively small equilibrium constants were expected
to evolve, we used the least polar solvent applicable, which was
toluene with practical solubilities for all reaction components at
all temperatures.
0.0034
0.0035
0.0036
0.0037
0.0038
0.0039
The K(
and K
values obtained for the hydrogen bonded
ðDR1Þ
D
d)
1 / T (K-1)
adducts of isopropanol with 1 showed similar dependencies in
temperature and the van’t Hoff plots of lnK vs. 1/T (Eq. (1), (2)) pro-
duced satisfactory linear regressions (Fig. 7). The thermodynamic
Fig. 6. van’t Hoff plots of the two hydrogen bonding equilibria K1 and K2 of 1
(0.0173 mol/L) and TMP (0.0567 mol/L) in hexane forming the adducts 3a (d) and
3b (j) according to K1 and K2 of Scheme 2. The equilibrium constants were
calculated from the concentrations of the equilibrium species derived from the
data calculated from K(
and K values were similar (
D
H(
Dd)
=
ðDR1Þ
D
d)
ꢁ2.3 kcal/mol;
D
S( d) = ꢁ11.7 e.u. and
D
H D = ꢁ2.9 kcal/mol;
D
ð R1Þ
deconvoluted
m
(WH) bands of Fig. 5.
D
ð
S D = ꢁ14.6 e.u.) presumably indicating that involvement of
R1Þ

´
387
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
by the IR method (2.8 kcal/mol). The studies of hydrogen-bonding
interactions of rhenium hydrides reported earlier with fluorinated
alcohols in hexane and in toluene showed similar qualitative
agreement [27].
K
(ΔR1)
2.0
1.5
The small negative free energy difference
mol of the IR derived results for 1 and TMP was in favour of K1 with
3a over K2 with 3b. The K values of 7a were determined to be
D
G of ca. ꢁ0.9 kcal/
K(Δδ)
1.0
ðDR1Þ
3a,b
less than 1 L/mol in the temperature range of 193–223 K, which
goes along with the very small positive free energy differences
revealing that the formation of any hydrogen bonded adduct of 1
with the weak acid isopropanol is generally less favourable. The
0.5
K
(ΔR1)
K(Δδ)
0.0
twice higher negative value of DH D of 3a than of 7a (Table 1)
ð R1Þ
confirmed much stronger hydrogen-bonding interaction between
1 and the more acidic TMP. It was also worth analyzing the entropy
effects of the hydrogen bonding adducts, although these are ex-
7a,b
-0.5
-1.0
pected to possess greater errors. The
IR spectroscopy has a slightly smaller negative value than
tained by the T1 NMR spectroscopic measurements ( R1) (Table
1). A lower S value was determined by the IR analyses in hexane,
D
S value of 3a observed by
D
S ob-
D
0.0042
0.0044
0.0046
0.0048
0.0050
0.0052
0.0054
D
1/T (K-1)
which might be explained by a solvent effect and a somewhat dif-
ferent self-association process of the free TMP in hexane and tolu-
ene solution.
Fig. 7. van’t Hoff plots lnK vs. 1/T for the adduct equilibria of 1 with TMP or
isopropanol forming 3a,b or 7a,b in d8-toluene. The K values were derived from
NMR titrations observing the chemical shifts of HW, (
D
d) or the changes in the
The dihydrogen bond distance r(HꢀꢀꢀH) [Å] obtained from
excess relaxation rates of HW (
protic donors.
DR1) in dependence of the concentration of the
r(HꢀꢀꢀH) = 5.817(
m D
R1(min))ꢁ1/6 are 2.1 Å for the dihydrogen bond-
ꢀ
ing between 1 and isopropanol (7a) and 1.9 Å for the one between
1 and TMP (3a) (Table 1) showing a difference in the contact dis-
tance of 0.2 Å. The 1.9 Å intermolecular bonding of 3a expresses
a relatively strong interaction, while the 2.1 Å contact of 7a is
weak, falling into the upper range for dihydrogen bonding interac-
tions (1.7–2.2 Å) [1,2,29].
the hydrogen bonding equilibrium with ONO leading to 7b is very
small and practically non-existent.
In the case of the adduct formation of 1 with TMP producing 3a
and 3b, the van’t Hoffs plots showed also good linear regressions,
however, with quite different slopes for K(
and K
(Fig. 7)
D
d)
ðDR1
Þ
3. Conclusion
and consequently with large differences in the calculated thermo-
dynamic data S( d) = ꢁ1.4 e.u. and
H( d) = ꢁ0.8 kcal/mol;
H D = ꢁ5.8 kcal/mol;
(D
D
D
D
Protonation of 1 with various ROH protic donors produced orga-
nyloxide complexes with evolution of H2. Experimental evidence
was provided for intermolecular dihydrogen bonding between 1
and the protic donors, which might precede the H2 evolution pro-
cess. In general hydrogen-bonding interactions are equilibria and
therefore dependent on temperature and type of solvent, the
ROH concentration, the ROH acid strength and to some extent also
on steric effects. IR experiments at low temperatures revealed in
the case of the protic donor TMP the expected dihydrogen bonding
to HW of 1 (3a) and another competitive equilibrium process of
hydrogen bonding to the ONO group to form 3b. The interaction
D
D
S D = ꢁ22.9 e.u.). This is interpreted in
ð
R1
Þ
ð R1Þ
terms of the appearance of stronger hydrogen bonding to ONO
implicitly contained in the d measurements, but not in the R1
D
D
measurements. The hydrogen bonding equilibrium of the proton
of TMP with the ONO group of 3b (also evidenced from the IR spec-
troscopic studies) interferes strongly in the averaging of the chem-
ical shifts of HW and the calculated K(
values are indeed a
D
d)
mixture of both the K1 and K2 values according to Scheme 2. The
values represent only dihydrogen bonding interaction with
K
ðDR1Þ
the hydride ligand of 1 to form 3a.
As described in the previous chapter the
DH of the interaction of
enthalpies of 3a observed in the IR (
DH1) and NMR titrations
1 with TMP was also determined by IR spectroscopy in the temper-
ature range of 253–293 K (Table 1). Equilibrium constants were di-
rectly calculated from the scaled band intensities. The only
difference between the IR and the NMR experiments was the sol-
vent used: hexane vs. toluene, as unpolar solvents these apparently
(
D
H D Þ) gave the same value of ꢁ5.8 kcal/mol demonstrating
ð
R1
excellent agreement between these different spectroscopic meth-
odologies. From NMR relaxation time measurements of the hydride
ligand of 1 in d8-toluene, a shorter dihydrogen bonding distance
value was derived for 3a than for 7a demonstrating the pKa depen-
dence of the dihydrogen bonding interactions and stressing the
involvement of such species as intermediates in the protonation
of hydridic transition metal hydrides.
did not cause great differences. Thus, the
D
H values obtained for
the formation of 3a by both the IR and the
DR1 method agree very
well showing exactly the same value of ꢁ5.8 kcal/mol (Table 1).
The H for the interaction in 3b could reliably be obtained only
D
4. Experimental
Table 1
IR and NMR derived
4.1. General considerations
D
H and
D
S values and HꢀꢀꢀH distance in hexane (IR) or d8-toluene
(NMR) for the hydrogen bonding adducts 3a,b and 7a,b with reference to Figs. 6 and
All reactions and manipulations were performed under an
atmosphere of dry nitrogen using conventional Schlenk techniques
or a glove box. Solvents were dried by standard methods and
freshly distilled before use. Reagents of commercial quality were
obtained from commercial suppliers and used as received. NMR
spectra were recorded on the following spectrometers: Varian
Gemini-200 instrument; 1H at 199.98 MHz, 13C at 50.29 MHz, 19F
7.
Equilibrium and method
(7a)
K(IR) (3a)
K(IR) (3b)
D
H (kcal/mol)
D
S (e.u.)
r(HꢀꢀꢀH) (Å)
2.1
K
ðDR1Þ
ꢁ2.9 0.2
ꢁ5.8 0.1
ꢁ2.8 0.1
ꢁ5.8 0.4
ꢁ14.6 1.0
ꢁ15.3 0.3
ꢁ5.8 0.3
ꢁ22.9 1.9
K (3a)
ðDR1Þ
1.9

´
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
388
at 188.15 MHz, 31P at 80.95 MHz. Varian Gemini-300 instrument;
1H at 300.08 MHz, 13C at 75.46 MHz, 19F at 282.33 MHz, 31P at
121.47 MHz. Bruker DRX-500 instrument; 1H at 500.13 MHz, 13C
at 125.23 MHz, 31P at 202.51 MHz. Chemical shifts d (1H) and d
for C12H29F3NO3P3W: C, 25.32; H, 5.14; N, 2.46. Found: C, 25.40; H,
5.04; N, 2.44%.
4.5. Preparation of mer-W(CO)(NO)(OCH2Ph)(PMe3)3 (5)
(
(
13C) were recorded rel. to the solvent, d (19F) rel. to CFCl3, and d
31P) rel. to H3PO4. All NMR spectra were recorded at room temper-
Compound 1 (21 mg, 0.044 mmol) was dissolved in C6D6
ature unless otherwise stated. The two trans phosphines are re-
ferred to as ‘‘PMe3 trans”. IR spectra: Biorad FTS-45 and FTS-3500
instruments. Mass spectra: Finnigan-MAT-8400 spectrometer;
FAB spectra in 3-nitrobenzyl alcohol matrix. Elemental analyses:
Leco CHN(S)–932 instrument. mer-W(CO)(H)(NO)(PMe3)3 1 was
prepared according to a published procedure [64].
(0.7 mL), and benzyl alcohol (5 lL, 0.066 mmol) was added. After
1 d NMR monitoring at 60 °C the reaction was complete and the
solvent removed in vacuo. The residue was extracted with pentane
(0.5 mL), the solution filtered, and orange 5 crystallized at –30 °C.
Yield: 20 mg (78%). IR (cmꢁ1, pentane): 1904 (CO), 1572 (NO). 1H
NMR (300.0 MHz, C6D6): d 7.36 (m, 2H, Ph), 7.26 (m, 2H, Ph),
7.10 (m, 1H, Ph), 4.86 (s, 2H, –OCH2–), 1.29 (m, 18H, 2 PMe3 trans),
1.13 (d, 2JHP = 7 Hz, 9H, 1 PMe3).13C{1H} NMR (125.2 MHz, C6D6): d
4.2. Preparation of mer-W(CO)(NO)(OC6H5)(PMe3)3 (2)
2
2
229.2 (dt, JCP(cis zu CO) = 6 Hz, JCP(trans zu CO) = 54 Hz, CO), 149.4 (s,
Ph), 127.7 (s, Ph), 125.7 (s, Ph), 125.5 (s, Ph), 75.8 (dt, 3JCP(cis zu CO)
1 Hz, JCP(trans zu CO) = 8 Hz, –OCH2–), 18.2 (m, 2 PMe3 trans), 17.2
=
Orange
2 was prepared via the reaction of 1 (13 mg,
3
0.028 mmol) with phenol (2.7 mg, 0.028 mmol) in toluene-d8
(0.7 mL) at 60 °C stirring for 4 h. Yield: 11 mg (70%). IR (cmꢁ1, pen-
tane): 1910 (CO), 1584 (NO). 1H NMR (300.0 MHz, C6D6): d 7.24 (m,
2H, Ph), 6.67 (m, 3H, Ph), 1.22 (m, 18H, 2 PMe3 trans), 1.11 (d,
2JHP = 7 Hz, 9H, 1 PMe3). 13C{1H} NMR (125.2 MHz, C6D6): d 228.3
(dt, JCP = 20 Hz, JCP = 3 Hz, 1 PMe3). 31P{1H} NMR (121.5 MHz,
C6D6): d ꢁ25.2 (m mit Satelliten, 2 PMe3 trans), ꢁ26.2 (m mit Sat-
elliten, 1 PMe3). MS (EI, 70 eV): m/z 577 [M+], 549 [M+–CO], 501
[M+–PMe3], 473 [M+–CO–PMe3], 321 [M+–CO–3 PMe3], 291 [M+–
CO–3 PMe3–NO]. Anal. Calc. for C17H34NO3P3W: C, 35.37; H, 5.94;
N, 2.43. Found: C, 35.12; H, 5.54; N, 2.35%.
1
3
2
(dt, JCP(trans to CO) = 55 Hz, CO), 168.2 (m, Ph), 129.6 (s, Ph), 119.5
(s, Ph), 115.2 (m, Ph), 18.3 (m, 2 PMe3 trans), 17.1 (dt, 1JCP = 21 Hz,
3JCP = 3 Hz, 1 PMe3). 31P{1H} NMR (121.5 MHz, C6D6): d ꢁ23.0
4.6. Preparation of mer-W(CO)(NO)(OCH3)(PMe3)3 (6)
2
1
(d with satellites, JPP = 19 Hz, JPW = 289 Hz, 2 PMe3 trans), ꢁ26.9
2
1
(t with satellites, JPP = 19 Hz, JPW = 239 Hz, 1 PMe3). FAB-MS:
m/z 563 [M+], 535 [M+ꢁCO], 487 [M+ꢁPMe3]. Anal. Calc. for
C16H32NO3P3W: C, 34.12; H, 5.73; N, 2.49. Found: C, 34.51; H,
5.48; N, 2.34%.
mer-W(CO)(H)(NO)(PMe3)3 1 (16 mg, 0.034 mmol), was dis-
solved in C6D6 (0.7 mL), and methanol (4
lL, 0.099 mmol) was
added. After 1 d at 60 °C the solvent was removed in vacuo. The res-
idue was extracted with pentane (0.5 mL), the solution filtered, and
orange 6 crystallized at –30 °C. Yield: 14 mg (82%). IR (cmꢁ1, pen-
tane): 1904 (CO), 1569 (NO). 1H NMR (300.0 MHz, C6D6): d 3.92 (m,
4.3. Preparation of mer-W(CO)(NO)(O(3,4,5-Me3C6H2))(PMe3)3 (3)
2
3H, -OCH3), 1.36 (m, 18H, 2 PMe3 trans), 1.11 (d, JHP 7 Hz, 9H, 1
2
PMe3). 13C{1H} NMR (125.2 MHz, C6D6): d 229.2 (dt, JCP(trans to CO)
To a solid mixture of 1 eq of 1 (0.030 g, 0.064 mmol) and 1.1 eq
of 3,4,5-trimethylphenol (0.0095 g, 0.070 mmol) was added 1 ml of
toluene-d8. After 5 h at 60 °C the reaction was complete (NMR
monitoring). The solvent was removed in vacuum and the residue
was recrystallized from THF at –30 °C to give an orange–yellow so-
lid of 3. Yield: 0.033 g (85%). IR (cmꢁ1, pentane): 1582 (NO); 1908
(CO). 1H NMR (200 MHz, THF-d8, 25 °C): d 1.30 (m, 18H, 2PMe3,
trans), 1.10 (d, 9H, 1PMe3), 6.70 (s, 2H, Ph), 2.01 (s, 6H, 2CH3 from
Ph), 1.91 (s, 3H, CH3 from Ph). 31P{1H} NMR (80.9 MHz, THF-d8,
2
3
= 55 Hz, JCP(cis
= 6 Hz, CO), 62.2 (dt, JCP(trans
= 9 Hz,
to CO)
to CO)
3JCP(cis
= 1 Hz, –OCH3), 18.2 (m, 2 PMe3 trans), 17.1 (dt,
to CO)
1JCP = 20 Hz, JCP = 3 Hz, 1 PMe3). 31P{1H} NMR (121.5 MHz, C6D6):
d ꢁ24.4 (m with satellites, 2 PMe3 trans), -26.7 (m with satellites,
1 PMe3). EI-MS: m/z 501 [M+], 473 [M+–CO], 397 [M+–CO–PMe3].
For this compound no satisfactory elemental analysis could be
obtained.
3
2
1
25 °C): d ꢁ30.3 (t with satellites JPP = 19.9 Hz, JPW = 300 Hz, 1
4.7. Preparation of mer-W(CO)(NO)[OCHMe2](PMe3)3 (7)
2
1
PMe3), ꢁ26.0 (d with satellites, JPP = 19.9 Hz, JPW = 350 Hz, 2
PMe3, trans). 13C{1H} NMR (50.3 MHz, THF-d8, 25 °C): d 228.6 (dt,
Compound 1 (28 mg, 0.058 mmol), was dissolved in C6D6
(0.7 mL), and isopropanol (18 lL, 0.232 mmol) was added. After 3
2JCP(trans to CO) = 56 Hz, JCP(cis to CO) = 6 Hz, CO), 166.5 (s, Ph), 135.2
2
(s, Ph), 119.7 (s, Ph), 118.0 (s, Ph), 21.0 (s, CH3 from Ph), 14.2 (s,
d at 60 °C the solvent was removed in vacuo. The residue was ex-
tracted with pentane (0.5 mL), the solution filtered, and yellow–or-
ange 7 crystallized at –30 °C. Yield: 21 mg (68%). IR (cm-1,
pentane): 1905 (CO), 1565 (NO). 1H NMR (300.0 MHz, C6D6): d
3.73 (sept, 3JHH = 6 Hz, 1H, –OCHMe2), 1.37 (m, 18H, 2 PMe3 trans),
1
3
CH3 from Ph), 18.6 (td, JCP = 12.3 Hz, JCP = 2.1 Hz, 2 PMe3, trans),
1
3
17.6 (dt, JCP = 20.0 Hz, JCP = 3 Hz, PMe3). Anal. Calc. for
C19H38NO3P3W: C, 37.69; H, 6.28; N, 2.31. Found: C, 37.90; H,
6.23; N, 2.38%.
2
3
1.09 (d, JHP = 6 Hz, 9H, 1 PMe3), 1.05 (d, JHH = 6 Hz, 6H, –CH3).
13C{1H} NMR (125.2 MHz, C6D6): d 229.4 (m, CO), 70.9 (m,
–OCHMe2), 28.8 (s, –CH3), 18.3 (m, 2 PMe3 trans), 16.8 (m, 1
PMe3). 31P{1H} NMR (121.5 MHz, C6D6): d –27.9 (m mit Satelliten,
3 PMe3). (EI-MS): m/z 529 [M+], 501 [M+–CO], 425 [M+–O–PMe3],
349 [M+–CO–2 PMe3]. Anal. Calc. for C13H34NO3P3W: C, 29.51; H,
6.48; N, 2.65. Found: C, 29.45; H, 6.05; N, 2.68%.
4.4. Preparation of mer-W(CO)(NO)(OCH2CF3)(PMe3)3 (4)
Yellow–orange 4 was prepared using the same procedure as for
3, reacting 1 (20 mg, 0.042 mmol) and 2,2,2-trifluoroethanol (9.3
l
L, 0.127 mmol) in toluene-d8 (0.8 mL) at 60 °C for 1 d. Yield:
15 mg (62%). IR (cmꢁ1, pentane): 1908 (CO), 1580 (NO). 1H NMR
3
(300.0 MHz, C6D6): d 3.86 (q, JHF = 10 Hz, 2H, –OCH2–), 1.26 (m,
18H, 2 PMe3 trans), 1.03 (d, JHP = 6 Hz, 9H, 1 PMe3). 13C{1H}
4.8. Quantitative NMR pursuits of the reaction of 1 with protic donors
2
1
NMR (125.2 MHz, C6D6): d 229.0 (m, CO), 128.1 (q, JCF = 282 Hz,
2
3
3
-CF3), 71.8 (dtq, JCF = 31 Hz, JCP(trans
= 8 Hz, JCP(cis
=
Kinetic measurements were carried out on a Varian-Gemini-
200 spectrometer (80.9 MHz, 31P{1H}). 20 mg of 1 (0.0425 mmol)
was dissolved in 0.7 ml of toluene-d8 and 8.0 mg of phenol
(0.085 mol/L); 11.6 mg of 3,4,5-trimethylphenol (0.085 mol/L);
to CO)
to CO)
2 Hz, –OCH2–), 18.2 (m, 2 PMe3 trans), 16.8 (m, 1 PMe3). 19F NMR
(282.3 MHz, C6D6): d -78.7 (t, JFH = 10 Hz, –CF3). 31P{1H} NMR
3
(202.5 MHz, C6D6): d ꢁ24.6 (m, 2 PMe3 trans), -25.4 (m, 1 PMe3).
EI-MS: m/z 569 [M+], 541 [M+–CO], 465 [M+–CO–PMe3]. Anal. Calc.
9.3 lL of trifluoroethanol (0.1275 mmol); 13.2 lL of benzylalcohol

´
389
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
(0.1275 mmol); 5.2
l
L of methanol (0.1275 mmol) and 9.8
l
L of
4.10. Qualitative analysis of hydrogen bonding of 1 with isopropanol
isopropanol (0.1275 mmol) were placed in a Young tap NMR tube.
The rate of formation of the alkoxide complexes 2–7 and H2 evolu-
tion were monitored by integration of the phosphine doublets and
the triplet resonances of hydride 1 at –29.8 and –35.1 ppm in the
31P{1H} NMR spectra at every 8.5 min (2), 9.3 min (3), 32.5 min
(4), 32.5 min (5), 32.5 min (6) and 97.6 min (7) at 60 °C. The reac-
tions of formation of 2–7 were over after 4 h (2), 5 h (3), 20 h (4),
24 h (5), 24 h (6) and 72 h (7) when 1 was completely consumed.
Plot of the decay of [1] vs. time t allowed exponentional fitting
with a first order law. From the slope of the linear plot of lnc vs.
time t, the rate constant k was determined. The half-life times of
and 3,4,5-trimethylphenol using VT 1H NMR spectroscopy
1H and T1 NMR experiments were obtained on a Bruker DRX-
500 spectrometer (500 MHz, 1H). The inversion-recovery method
(180-s-90) was used to determine T1 relaxation times. The calcula-
tion of the relaxation times was made using the nonlinear three-
parameter fitting routine of the spectrometer. In each experiment,
the waiting period was 10 times the expected relaxation time and
12 variable delays were employed. The duration of the pulses was
controlled at every studied temperature. Dihydrogen bonding is
solvent dependent and for this reason toluene-d8 was used as
one of the least polar solvents which still provided good solubility
of all reaction components especially at low temperature.
the reactions t1/2 were calculated based on the expression t1/2
ln2/k [83].
=
4.11. Equilibrium analysis of the dihydrogen bonded adducts 7a
(WHꢀꢀꢀHOCHMe2) and 7b (W–NOꢀꢀꢀHOCHMe2) obtained from 1 and
isopropanol
4.9. Quantitative IR analysis of hydrogen bonding of 1 with 3,4,5-
trimethylphenol in the m(WH) region
For these studies, the Merlin Software [84] was used to carry
out quantitative analysis using the Lambert–Beer law. The Lam-
bert–Beer law is defined by the linear relationship between the
absorption and the concentration of the sample:
A solution of 1 (20 mg of 1 in 0.7 ml of toluene-d8, ct(WH) =
0.061 mol/L) was prepared in an NMR tube below 223 K and the
chemical shifts d(WH) and the temperature-dependent minimum
relaxation time T1(min)(WH) of the hydride ligand of 1 were deter-
mined in the temperature range of 193–228 K. Then, five different
solutions of 20 mg of 1 in 0.7 ml of toluene-d8 (ct(WH) = 0.061 mol/
L) with various concentrations of isopropanol (c(alc) = 0.0671,
0.1080, 0.1921, 0.2446 and 0.3019 mol/L) at temperature below
223 K were prepared in the NMR tubes sealed with the Teflon caps.
The chemical shifts d(eq) and the values T1(min)(eq) of the hydride
ligand of 1 were determined at every 5 K from 193 to 228 K. VT
NMR titrations were used in order to extract and compare equilib-
Ak ¼ ek ꢀ b ꢀ c
where Ak is the absorbance value of the sample at specific wave-
length, in our case at 1618 cmꢁ1
(m(W–H) band), ek is the molar
absorptivity at the wavelength (L molꢁ1 cmꢁ1), b is path length
through the sample (b = 0.1 cm) and c is the concentration of the
sample (mol/L). The calibration was carried out with linear curves
of known concentrations of 1 in hexane (c = 0.0030, 0.0064, 0.0110,
0.0191, 0.0232, 0.0364 and 0.0490 mol/L) in dependence of the
rium constants K(
and K
from the given changes in the
D
d)
ð
DR1
Þ
measured peak areas (or absorbances) of the selected
m(W–H)
chemical shift differences (Dd = d(eq) - d(WH)) and from the given
band at 1618 cmꢁ1 in the temperature range of 253–293 K. Further-
more, quantitative IR experiments were carried out involving a mix-
ture of 4.1 mg of 1 (0.0173 mol/L) and 3.9 mg (0.0567 mol/L) of
3,4,5-trimethylphenol in 0.5 ml of hexane. The temperature was
varied and the spectra were measured every 10 K starting from
253 K to 293 K. In the IR spectra the three overlapping bands of
the dihydrogen bonded species C6H2(CH3)3OHꢀꢀꢀH–W(CO)(N-
O)(PMe3)3 (3a) at 1599 cmꢁ1, of the free hydride 1 at 1618 cmꢁ1
and of the dihydrogen bonding species C6H2(CH3)3OHꢀꢀꢀON–
W(CO)(H)(PMe3)3 (3b) at 1625 cmꢁ1 were deconvoluted using a
programme with a Lorentzian line function [85] and then the equi-
librium concentrations of the species [3a], [1] and [3b] were deter-
mined using the calibration curves of 1 at each temperature. In
these calculations we used for the two adducts 3a and 3b the same
extinction coefficients for the W–H bands and we have good indica-
tion that the error stays within acceptable limits. The initial concen-
tration of 1 should be the sum of concentrations of all the species 1,
3a and 3b in equilibrium. Based on the peak areas calibrated to hy-
dride 1, the overall deviations of the sum of concentrations from
100% are in the range of 3.9–8.7%. The equilibrium concentrations
of 3,4,5-trimethylphenol were calculated from the initial concentra-
changes in the excess relaxation rates (DR1(min) = 1/T1(min)(eq)–
1/T1(min)(WH)) by curve fitting.
The equilibrium constants K(
and K Þwere calculated from
ð R1
D
d)
D
Eqs. (3)–(6).
K ¼ cðaddÞ=cðWHÞ ꢀ cðalcÞ
ð3Þ
ð4Þ
dðeqÞ ¼ dðWHÞ þ ðdðaddÞ ꢁ dðWHÞÞ ꢀ XðaddÞ
1=T1ðminÞðeqÞ ¼ 1=T1ðminÞðaddÞ ꢀ XðaddÞ þ 1=T1ðWHÞ ꢀ XðWHÞ
ð5Þ
1=T1ðminÞðaddÞ ¼ ð1=T1ðminÞðeqÞ ꢁ 1=T1ðWHÞ ꢀ XðWHÞÞ=XðaddÞ
ð6Þ
c(add), c(WH) and c(alc) ofEq. (3) are the equilibrium concentrations
of 7a,b or 3a,b, 1 and the protic donors, respectively, and K is the
equilibrium constant. d(eq) of Eq. (4) is the averaged equilibrium
chemical shift of d(WH) and d(add), d(WH) and d(add) are the chem-
ical shifts of 1 and 7a,b or 3a,b, respectively. ct(WH) and ct(alc) are
the initial concentrations of 1 and of the alcohols. X(add) = (0.5/
ct(WH))ꢀct(WH) + c t(alc) + 1/K–(c t(WH) + c t(alc) + 1/K)2
–
tion of C6H2(CH3)3OH by the relation: [C6H2(CH3)3OH]
=
4ct(WH)ꢀct(alc)1/2 is the molar fraction of 7a,b or 3a,b, which is ob-
tained from the curve fitting of the experimental data of Eq. (3). T1
(min)(eq) ofEq. (5) is the equilibrium-averaged temperature-depen-
dent minimum relaxation time; T1(WH) is the relaxation time of 1
at the temperature of T1(min); T1(min)(add) and T1(min)(WH) are
the temperature–dependent minimum relaxation times of the hy-
dride ligands of 7a or 3a and 1, respectively. X(WH) and X(add)
are the molar fractions of complexes 1 and 7a or 3a at the temper-
ature of T1(min)(eq).
[C6H2(CH3)3OH]initial – [3a]–[3b]. The equilibrium constants K1
and K2 were calculated by the equations:
K1 ¼ ½3aꢂ=½1ꢂ½C6H2ðCH3Þ3OHꢂ
ð1Þ
K2 ¼ ½3bꢂ=½1ꢂ½C6H2ðCH3Þ3OHꢂ
ð2Þ
The calculated equilibrium constants K1 and K2 were fitted by
van’t Hoff linear regressions. From the van’t Hoff linear plots of
lnK1 and lnK2 vs. 1/T, the enthalpies
D
H1 and
DH2 with errors of
7% and 4% and the entropies
5% were determined.
D
S1 and
D
S2 with errors of 9% and
Eqs. (4) and (5) represent the basic equations of the NMR titra-
tion experiments [12,27,86], which can be approached by curve fit-

´
390
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
Table 2
ting of the curves of d(eq) and 1/T1(min)(eq) vs. the initial alcohol
concentrations ct(alc). The K values were obtained by nonlinear fits
using a Levenberg–Marquardt algorithm.[45] For this purpose, Eq.
(7) (modified Eq. (4)) [87] was used (see below) where Q = 0.5 ꢀ
(S–a), S = (a2 + 4 ꢀ K ꢀct (WH))1/2 and a = K ꢀ (ct(WH) – ct (alc)) + 1.
Summary of crystallographic data for complex 2.
Compound
2
Empirical formula
C16H32NO3P3W
563.19
183(2)
Formula weight (gꢀmolꢁ1
)
Temperature (K)
dðeqÞ ¼ ðdðWHÞ þ Q ꢀ dðaddÞÞ=ð1 þ QÞ
In this way, the association constants K( d) and K
ð7Þ
for the ad-
Wavelength (Å)
Crystal system, space group
a (Å)
b (Å)
c (Å)
0.71073
Monoclinic, P 21/n
11.4798(8)
13.9009(8)
15.1070(10)
90
D
ðDR1
Þ
duct formation reaction were determined at different tempera-
tures and then X(add) and X(WH) were calculated by solving Eq.
(3).
a
(°)
b (°)
107.912(8)
90
2293.9(3)
4, 1.631
1
þ HORðR ¼ CHðCH3Þ2; 3; 4; 5 ꢁ Me3C2H2Þ ! 7a; b; 3a; b
c (°)
x
ctðWHÞꢁx
Volume (Å3)
ctðalcÞꢁx
Z, density (calcd) (Mgꢀmꢁ3
)
Abs coefficient (mmꢁ1
Crystal size (mm3)
)
5.258
K ¼ cðaddÞ=cðWHÞ ꢀ cðalcÞ ¼ x=ðctðWHÞ ꢁ xÞ ꢀ ðctðalcÞ ꢁ xÞ
XðaddÞ ¼ x=ctðWHÞ
ð8Þ
ð9Þ
0.26 ꢃ 0.26 ꢃ 0.17
Numerical
0.487/0.342
0.0351, 0.0707
0.0624, 0.0734
Absorption correction
Max/min transmission
Final R1 and wR2 indices [I > 2
R1 and wR2 indices (all data)
r(I)]
XðWHÞ ¼ 1 ꢁ XðaddÞ
ð10Þ
c(add) = x is calculated solving Eq. (8) and X(add) and X(WH) are
calculated from Eq. (9) and (10). Then, based on Eqs. (4) and (5),
d(add) and 1/T1(min)(add) values were calculated.
From the determined T1(min) values of the averaged hydride
resonance and subsequently from their excess relaxation rates
exposed at a constant time of 1.50 min/image. The total exposure
and read-out time was 16 h. The crystal-to-image distance was
set to 50 mm (hmax = 30.31°). The / rotation mode was selected
for the / increment of 1.2° per exposure. A total of 8000 reflections
D
R1(min), the distance to the OH protons could be calculated
[12,27,86,88–90]. Distances r(HꢀꢀꢀH) in [Å] can be obtained from
Eq. (11) [12,27,91–95] where is the NMR frequency in MHz and
R1(min) = 1/T1(min)(add) – 1/T1(min)(WH).
with I > 6r(I) was selected out of the whole limiting sphere for the
cell parameter refinements. A total of 26902 was collected of
which 6588 were unique (Rint = 0.0544). The intensities were cor-
rected from Lorentz and polarization factors and numerical absorp-
tion corrections based on 8 indexed crystal faces were also applied.
The Patterson method was used to solve the crystal structures of 2
by applying the software options of the program shelxs-97 [97].
The structure refinements were performed with the program shel-
xl-97 [97]. The program PLATON [98] was used to check the result of
the X-ray analysis and the program ORTEP [99] used to give a repre-
sentation of the structure. Details of crystallographic data and
refinements of complex 2 are summarized in Table 2.
m
D
rðH ꢀ ꢀ ꢀ HÞ ¼ 5:817ð
m
ꢀ
D
R1ðminÞÞꢁ1=6
ð11Þ
From van’t Hoff linear regressions of lnK(
and lnK vs. 1/T
ðDR1Þ
D
d)
the enthalpies
DH( d) and DH D with errors of 8.7% and 6.9% and
D
ð R1Þ
the entropies
determined.
D
S(
D
and DS D with errors of 7.7% and 6.8% were
ð R1Þ
d)
4.12. Equilibrium analysis of the dihydrogen bonded adducts 3a
(WHꢀꢀꢀHOC6H2(CH3)3) and 3b (W–NOꢀꢀꢀHOC6H2(CH3)3) between 1 and
3,4,5-trimethylphenol (TMP)
Acknowledgment
Compound 1 was dissolved in 0.5 ml of toluene-d8 (7 mg,
ct(WH) = 0.0297 mol/L) in an NMR tube below 223 K and the chem-
ical shifts d(WH) and T1(min)(WH) of the hydride ligand of 1 were
determined in the temperature range of 203–243 K. Under the
same conditions the chemical shifts d(eq) and the values
T1(min)(eq) of the hydride ligand of 1 were determined for the five
different solutions of 7 mg of 1 in 0.7 ml of toluene-d8 (ct(WH) =
0.0297 mol/L) with various concentrations of 3,4,5-trimethylphe-
nol (c(alc) = 0.0315, 0.0591, 0.0784, 0.1099 and 0.1342 mol/L) pre-
We thank the University of Zürich and the Swiss National Sci-
ence Foundation for financial support.
Appendix A. Supplementary material
CCDC 728847 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_request/cif. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/j.jorganchem.
2009.10.036.
pared in Young tap NMR tubes. Equilibrium constants K(
and
D
d)
K were determined in an analogous way as for the equilibrium
ðDR1Þ
analysis of 1 and isopropanol. The determined enthalpies
and H D were obtained with errors of 12.5% and 6.9%, while
the entropies
21.4% and 8.3%.
DH(
Dd)
D
ð
R1Þ
References
D
S(
and
d )
D
S D
were obtained with errors of
Þ
ð
R1
D
[1] R. Custelcean, J.E. Jackson, Chem. Rev. 101 (2001) 1963.
[2] L.M. Epstein, E.S. Shubina, Coord. Chem. Rev. 231 (2002) 165.
[3] E.S. Shubina, A.N. Krylov, A.Z. Kreindlin, M.I. Rybinskaya, L.M. Epstein, J.
Organomet. Chem. 465 (1994) 259.
4.13. X-ray diffraction study
[4] L.M. Epstein, E.S. Shubina, A.N. Krylov, A.Z. Kreindlin, M.I. Rybinskaya, J.
Organomet. Chem. 447 (1993) 277.
[5] E.S. Shubina, A.N. Krylov, N.V. Belkova, L.M. Epstein, A.P. Borisov, V.D. Mahaev,
J. Organomet. Chem. 493 (1995) 275.
[6] N.V. Belkova, A.V. Ionidis, L.M. Epstein, E.S. Shubina, S. Gründemann, N.S.
Golubev, H.-H. Limbach, Eur. J. Inorg. Chem. (2001) 2163.
[7] D. Milstein, J.C. Calabrese, I.D. Williams, J. Am. Chem. Soc. 108 (1986) 6387.
[8] R.C. Stevence, R. Bau, D. Milstein, O. Blum, T.F. Koetzle, J. Chem. Soc., Dalton
Trans. (1990) 1429.
A single crystal of 2 was mounted on top of a glass fibre using
polybutene oil as protecting agent, and immediately transferred
to the diffractometer. There the crystal was cooled to 183(2) K
by an Oxford cryogenic system. The determination of the unit cell
parameters and the collection of intensity data were performed
with an image plate detector system (Stoe IPDS diffractometer)
with graphite monochromated Mo Ka radiation using the Stoe
[9] A.J. Lough, S. Park, R. Ramachandran, R.H. Morris, J. Am. Chem. Soc. 116 (1994)
8356.
IPDS software version 2.92 (1999) [96]. A total of 167 images were

´
391
N. Avramovic et al. / Journal of Organometallic Chemistry 695 (2010) 382–391
[10] S. Park, R. Ramachandran, A.J. Lough, R.H. Morris, J. Chem. Soc., Chem.
Commun. (1994) 2201.
[52] A.A.H. van der Zeijden, H.W. Bosch, H. Berke, Organometallics 11 (1992)
2051.
[11] J.C. Lee, E. Peris, A.L. Rheingold, R.H. Crabtree, J. Am. Chem. Soc. 116 (1994)
11014.
[12] E.S. Shubina, N.V. Belkova, A.N. Krylov, E.V. Vorontsov, L.M. Epstein, D.G.
Gusev, M. Niedermann, H. Berke, J. Am. Chem. Soc. 118 (1996) 1105.
[13] N.V. Belkova, E.S. Shubina, A.V. Ionidis, L.M. Epstein, H. Jacobsen, A. Messmer,
H. Berke, Inorg. Chem. 36 (1997) 1522.
[14] E.S. Shubina, N.V. Belkova, E.V. Bakhmutova, L.N. Saitkulova, A.V. Ionidis, L.M.
Epstein, Russ. Chem. Bull. 47 (1999) 817.
[15] N.V. Belkova, E.S. Shubina, E.I. Gutsul, L.M. Epstein, I.L. Eremenko, S.E. Nefedov,
J. Organomet. Chem. 610 (2000) 58.
[16] L.M. Epstein, N.V. Shubina, E.V. Bakhmutova, L.N. Saitkulova, V.I. Bakhmutov,
A.L. Chistyakov, I.V. Stankevich, Inorg. Chem. 37 (1998) 3013.
[17] L.A. Vinogradova, A.Z. Kreindlin, L.A. Leites, I.T. Chizevskii, E.S. Shubina, L.M.
Epstein, Metalloorg. Khim. 3 (1990) 1192.
[18] S.G. Kazarian, P.A. Hamley, M. Poliakoff, J. Am. Chem. Soc. 115 (1993) 9069.
[19] S.G. Kazarian, P.A. Hamley, M. Poliakoff, J. Chem. Soc., Chem. Commun. (1992)
995.
[53] A.A.H. van der Zeijden, V. Shklover, H. Berke, Inorg. Chem. 30 (1991) 4393.
[54] A.A.H. van der Zeijden, D. Veghini, H. Berke, Inorg. Chem. 31 (1992) 5106.
[55] A.A.H. van der Zeijden, H. Berke, Helv. Chim. Acta 75 (1992) 513.
[56] A.A.H. van der Zeijden, H.W. Bosch, H. Berke, Organometallics 11 (1992)
563.
[57] H. Berke, P. Burger, Comments Inorg. Chem. 16 (1994) 279.
[58] D. Nietlispach, D. Veghini, H. Berke, Helv. Chim. Acta 77 (1994) 2197.
[59] F. Bäth, Ph.D. Thesis, University of Zürich, 1998.
[60] F. Furno, T. Fox, H.W. Schmalle, H. Berke, Organometallics 19 (2000) 3620.
[61] H. Jacobsen, H.W. Schmalle, A. Messmer, H. Berke, Inorg. Chim. Acta 306 (2000)
153.
[62] F.P. Liang, H. Jacobsen, H.W. Schmalle, T. Fox, H. Berke, Organometallics 19
(2000) 1950.
[63] F. Furno, T. Fox, M. Alfonso, H. Berke, Eur. J. Inorg. Chem. (2001) 1559.
[64] J. Höck, H. Jacobsen, H.W. Schmalle, G.R.J. Artus, T. Fox, J.I. Amor, F. Bäth, H.
Berke, Organometallics 20 (2001) 1533.
[65] A. Llamazares, H.W. Schmalle, H. Berke, Organometallics 20 (2001) 5277.
[66] X.Y. Liu, S. Bouherour, H. Jacobsen, H.W. Schmalle, H. Berke, Inorg. Chim. Acta
330 (2002) 250.
[20] M. Poliakoff, S.M. Howdle, S.G. Kazarian, Angew. Chem., Int. Ed. Engl. 34 (1995)
1275.
[21] E.S. Shubina, N.V. Belkova, L.M. Epstein, J. Organomet. Chem. 536 (1997) 17.
[22] R.H. Crabtree, J. Organomet. Chem. 557 (1998) 111.
[23] N.V. Belkova, M. Besora, L.M. Epstein, A. Liedós, F. Maseras, E.S. Shubina, J. Am.
Chem. Soc. 125 (2003) 7715.
[67] F.P. Liang, H.W. Schmalle, T. Fox, H. Berke, Organometallics 22 (2003) 3382.
[68] F.P. Liang, H.W. Schmalle, H. Berke, Inorg. Chem. 43 (2004) 993.
[69] Z. Chen, H.W. Schmalle, T. Fox, H. Berke, J. Chem. Soc., Dalton Trans. (2005)
580.
[24] N.V. Belkova, P.O. Revin, L.M. Epstein, E.V. Vorontsov, V.I. Bakhmutov, E.S.
Shubina, E. Collange, R. Poli, J. Am. Chem. Soc. 125 (2003) 11106.
[25] E.I. Gutsul, N.V. Belkova, M.S. Sverdlov, L.M. Epstein, E.S. Shubina, V.I.
Bakhmutov, T.N. Gribanova, R.M. Minyaev, C. Bianchini, M. Peruzzini, F.
Zanobini, Chem. Eur. J. 9 (2003) 2219.
[70] E. Bannwart, H. Jacobsen, F. Furno, H. Berke, Organometallics 19 (2000)
3605.
[71] D. Nietlispach, H.W. Bosch, H. Berke, Chem. Ber. 127 (1994) 2403.
[72] H. Jacobsen, K. Heinze, A. Llamazares, H.W. Schmalle, G. Artus, H. Berke, J.
Chem. Soc., Dalton Trans. (1999) 1717.
[26] E.V. Bakhmutova, V.I. Bakhmutov, N.V. Belkova, M. Besora, L.M. Epstein, A.
Liedós, G.I. Nikonov, E.S. Shubina, J. Tomàs, E.V. Vorontsov, Chem. Eur. J. 10
(2004) 661.
[27] A. Messmer, H. Jacobsen, H. Berke, Chem. Eur. J. 5 (1999) 3341.
[28] E.S. Shubina, A.N. Krylov, D.V. Muratov, A.A. Fil’chikov, L.M. Epstein, Russ.
Chem. Bull. 42 (1993) 1919.
[73] D.G. Gusev, A. Llamazares, G. Artus, H. Jacobsen, H. Berke, Organometallics 18
(1999) 75.
[74] H.-U. Hund, U. Ruppli, H. Berke, Helv. Chim. Acta 76 (1993) 963.
[75] D.G. Gusev, D. Nietlispach, I.L. Eremenko, H. Berke, Inorg. Chem. 32 (1993)
3628.
[76] Z. Yin, Ph.D. Thesis, University of Zürich, 2005.
[29] L.M. Epstein, N.V. Belkova, E.S. Shubina, in: M. Peruzzini, R. Poli, (Eds.),
Advances in Hydride Chemistry, Elsevier, 2001, p. 391.
[30] E. Peris, J.C. Lee, J.R. Rambo, O. Eisenstein, R.H. Crabtree, J. Am. Chem. Soc. 117
(1995) 3485.
[31] J.C. Lee, A. Rheingold, B. Muller, P.S. Pregosin, R.H. Crabtree, J. Chem. Soc.,
Chem. Commun. (1994) 1021.
[32] K. Abdur-Rashid, T.P. Fong, B. Greaves, D.G. Gusev, J.G. Hinman, S.E. Landau, A.J.
Lough, R.H. Morris, J. Am. Chem. Soc. 122 (2000) 9155.
[33] H.S. Chu, C.P. Lau, K.Y. Wong, Organometallics 17 (1998) 2768.
[34] Y. Musashi, S. Sakaki, J. Am. Chem. Soc. 122 (2000) 3867.
[35] A. Caballero, F.A. Jalón, B.R. Manzano, Chem. Commun. (1998) 1879.
[36] W. Yao, R.H. Crabtree, Inorg. Chem. 35 (1996) 3007.
[37] S. Aime, R. Gobetto, E. Valls, Organometallics 16 (1997) 5140.
[38] J. Wessel, J.C. Lee, E. Peris, G.P.A. Yap, J.B. Fortin, J.S. Ricci, G. Sini, A. Albinati,
T.F. Koetzle, O. Eisenstein, A.L. Rheingold, R.H. Crabtree, Angew. Chem., Int. Ed.
Engl. 34 (1995) 2507.
[39] G. Sini, O. Eisenstein, W. Yao, R.H. Crabtree, Inorg. Chim. Acta 280 (1998) 26.
[40] B.P. Patel, J. Wessel, W. Yao, J.C. Lee, E. Peris, T.F. Koetzle, G.P.A. Yap, J.B. Fortin,
J.S. Ricci, G. Sini, A. Albinati, O. Eisenstein, A.L. Rheingold, R.H. Crabtree, New J.
Chem. 21 (1997) 413.
[77] J. Höck, T. Fox, H.W. Schmalle, H. Berke, Chimia 53 (1999) 350.
[78] Z. Chen, Ph.D. Thesis, University of Zürich, 2004.
[79] A. Messmer, Ph.D. Thesis, University of Zürich, 1999.
[80] A.A.H. van der Zeijden, C. Sontag, H.W. Bosch, V. Shklover, H. Berke, D. Nanz, W.
von Philipsborn, Helv. Chim. Acta 74 (1991) 1194.
[81] E.M. Arnett, L. Joris, E. Mitchell, T.S.S.R. Murty, T.M. Gorrie, P.V.R. Schleyer, J.
Am. Chem. Soc. 92 (1970) 2365.
[82] M.C.S. Lopes, H.W. Thompson, Spectrochim. Acta 24A (1968) 1367.
[83] A.A. Frost, R.G. Pearson, in: Kinetics and Mechanism, second ed., John Wiley &
Sons, New York, London, 1961.
[84] (a) Win-IR, FTS 25 and FTS 40 Spectrometer Manuel, Technical Publications
Bio-Rad Labaratories, Cambridge.;
(b) MERLIN, Merlin Reference Manuel, Technical Publications Bio-Rad
Laboratories, Cambridge.
[85] (a) D.G. Cameron, D.J. Moffatt, J. Test. Evalut. 2 (1984) 78;
(b) W.-J. Yang, P. Griffiths, Comput. Enhanced Spectrosc. 3 (1983) 157;
(c) J.K. Kauppinen, D.J. Moffat, H.H. Mantsch, D.G. Cameron, Appl. Spectrosc. 3
(1981) 271.
[86] A. Abragam, The Principles of Nuclear Magnetism, Oxford University Press,
Oxford, 1973.
[41] B.P. Patel, W. Yao, G.P.A. Yap, A.L. Rheingold, R.H. Crabtree, Chem. Commun. 8
(1996) 991.
[42] G.J. Kubas, Acc. Chem. Res. 21 (1988) 120.
[87] W.H. Press, B.P. Flannery, S.A. Teukolsky, W.T. Vetterling, Numerical Recipes,
Cambridge University Press, Cambridge, 1986 (1988 reprint, Chapter 14.4).
[88] P.J. Desrosiers, L. Cai, Z. Lin, R. Richards, J. Halpern, J. Am. Chem. Soc. 113
(1991) 4173.
[89] R.H. Morris, Can. J. Chem. 74 (1996) 1907.
[90] D.G. Hamilton, R.H. Crabtree, J. Am. Chem. Soc. 110 (1988) 4126.
[91] R. Cammack, Nature 373 (1995) 556.
[92] R.T. Hembre, S. McQueen, J. Am. Chem. Soc. 116 (1994) 2141.
[93] K.-J. Haack, S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori, Angew. Chem., Int. Ed.
Engl. 36 (1997) 285.
[94] J.-S. Song, D.J. Szalda, R.M. Bullock, C.J.C. Lawrie, M.A. Rodkin, J.R. Norton,
Angew. Chem. 104 (1992) 1280.
[43] P.G. Jessop, R.H. Morris, Coord. Chem. Rev. 121 (1992) 155.
[44] F. Maseras, A. Lledós, E. Clot, O. Eisenstein, Chem. Rev. 100 (2000) 601.
[45] L.M. Epstein, E.S. Shubina, Ver. Bunsenges. Phys. Chem. 102 (1998) 359.
[46] E.S. Shubina, N.V. Belkova, E.V. Bakhumotova, L.N. Saitkulova, A.V. Ionidis, L.M.
Epstein, Izv. Akad. Nauk, ser. Khim. (1999) 846.
[47] L.M. Epstein, A.N. Krylov, E.S. Shubina, J. Mol. Struct. 322 (1994) 345.
[48] S.G. Kazarian, P.A. Hamley, M. Poliakoff, J. Am. Chem. Soc 115 (1993) 9069.
[49] W.T. Klooster, T.F. Koetzle, P.E.M. Siegbahn, T.B. Richardson, R.H. Crabtree, J.
Am. Chem. Soc. 121 (1999) 6337.
[50] E.S. Shubina, N.V. Belkova, E.V. Bakhmutova, E.V. Vorontsov, V.I. Bakhmutov,
A.V. Ionidis, C. Bianchini, L. Marvelli, M. Peruzzini, L.M. Epstein, Inorg. Chim.
Acta 280 (1998) 302.
[51] V.I. Bakhmutov, E.V. Bakhmutova, N.V. Belkova, C. Bianchini, L.M. Epstein, M.
Peruzzini, E.S. Shubina, E.V. Vorontsov, F. Zanobini, Can. J. Chem. 79 (2001)
475.
[95] J.-S. Song, D.J. Szalda, R.M. Bullock, C.J.C. Lawrie, M.A. Rodkin, J.R. Norton,
Angew. Chem., Int. Ed. Engl. 31 (1992) 1233.
[96] STOE-IPDS Software package. STOE & Cie, Darmstadt, Germany.
[97] G.M. Sheldrick, Acta Crystallogr., Sect. A64 (2008) 112.
[98] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.
[99] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.