Syntheses and structures of ytterbium(II) hydride and hydroxide complexes: Similarities and differences with their calcium analogues

Article abstract of DOI:10.1021/ic700403d

The Yb(II) hydride complex (DIPP-nacnac)YbH·THF (3-Yb, DIPP-nacnac = CH{(CMe)(2,6-iPr₂C₆H₃N)}₂) was prepared by a mild metathesis reaction of (DIPP-nacnac)Yb[N(SiMe ₃)₂]·THF with PhSiH₃. 3-Yb crystallizes as a dimer with bridging hydride ions, and its geometry is similar to that of the analogue calcium hydride complex (3-Ca). 3-Yb is well soluble in benzene and remarkably stable in solution at room temperature. Ligand exchange to the homoleptic Yb(II) complexes takes place at higher temperatures (3-Yb is less stable than the analogue 3-Ca). The soluble hydride complexes 3-Ca and 3-Yb are catalysts for the hydrosilylation of 1,1-diphenylethylene, but differences in the product distributions are observed. Slow hydrolysis of (DIPP-nacnac) Yb[N(SiMe₃)₂]·THF gave reduction of water and unidentified Yb(III) complexes. Fast hydrolysis at low temperature, however, resulted in the first Yb(II) hydroxide complex, (DIPP-nacnac)Yb(OH)·THF (4-Yb, 20% yield), which is a dimer with bridging hydroxide ions in the solid state. The crystal structure is isomorphous to that of the calcium analogue 4-Ca. 4-Yb is well soluble in benzene and considerably more stable against ligand exchange and formation of homoleptic species than 3-Yb.

Full text of DOI:10.1021/ic700403d

Inorg. Chem. 2007, 46, 5320−5326
Syntheses and Structures of Ytterbium(II) Hydride and Hydroxide
Complexes: Similarities and Differences with Their Calcium Analogues
Christian Ruspic, Jan Spielmann, and Sjoerd Harder*
Anorganische Chemie, UniVersita¨t Duisburg-Essen, UniVersita¨tsstrasse 5-7,
45117 Essen, Germany
Received March 2, 2007
The Yb(II) hydride complex (DIPP-nacnac)YbH·THF (3-Yb, DIPP-nacnac
) CH{(CMe)(2,6-iPr2C6H3N)}2) was
prepared by a mild metathesis reaction of (DIPP-nacnac)Yb[N(SiMe3)2]·THF with PhSiH3. 3-Yb crystallizes as a
dimer with bridging hydride ions, and its geometry is similar to that of the analogue calcium hydride complex
(3-Ca). 3-Yb is well soluble in benzene and remarkably stable in solution at room temperature. Ligand exchange
to the homoleptic Yb(II) complexes takes place at higher temperatures (3-Yb is less stable than the analogue
3-Ca). The soluble hydride complexes 3-Ca and 3-Yb are catalysts for the hydrosilylation of 1,1-diphenylethylene,
but differences in the product distributions are observed. Slow hydrolysis of (DIPP-nacnac)Yb[N(SiMe3)2]·THF gave
reduction of water and unidentified Yb(III) complexes. Fast hydrolysis at low temperature, however, resulted in the
first Yb(II) hydroxide complex, (DIPP-nacnac)Yb(OH)·THF (4-Yb, 20% yield), which is a dimer with bridging hydroxide
ions in the solid state. The crystal structure is isomorphous to that of the calcium analogue 4-Ca. 4-Yb is well
soluble in benzene and considerably more stable against ligand exchange and formation of homoleptic species
than 3-Yb.
Introduction
were only reported recently. This backlog in Ln(II) chemistry
is likely due to two factors: (i) Their much larger ionic radii
make it hard to saturate the coordination spheres. (ii) Their
much weaker and more ionic bonding of ligands results in
lower stability and a more pronounced tendency toward
ligand exchange (e.g., Schlenk equilibria).
The organometallic chemistry of the lanthanides (Ln )
lanthanide) has witnessed a rapid progress over the last
decades.1 This development is largely due to the enormous
potential in catalysis2 and new materials.3 Although this is
particularly true for the lanthanides in their most stable
oxidation state III, the chemistry of Ln(II) complexes is much
less developed. Despite extensive knowledge on cyclopen-
tadienyl-Ln(II) chemistry,1i-j very few alkyl complexes have
been published4 and the first aryl5 and benzyl6 complexes
The latter problems are very similar to those encountered
in the chemistry of the heavier alkaline-earth metals, and
consequently, parallels between these two research areas have
been found: e.g., there is ample evidence for the striking
similarity in the chemistry of Ca2+ and Yb2+.6,7 Despite
similarities also remarkable differences have been found.6
* To whom correspondence should be addressed. E-mail: sjoerd.harder@
uni-due.de.
We recently introduced synthetic pathways to and struc-
tures of two new functional groups in calcium chemistry:
(1) (a) Edelmann, F. T. Angew. Chem. 1995, 107, 2647; Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2466. (b) Eaborn, C.; Izod, K.; Smith, J. D.
J. Organomet. Chem. 1995, 500, 89. (c) Evans, W. J. New J. Chem.
1995, 19, 525. (d) Hogerheide, M. P.; Boersma, J.; van Koten, G.
Coord. Chem. ReV. 1996, 155, 87. (e) Anwander, R.; Herrmann, W.
A. Top. Curr. Chem. 1996, 179, 1. (f) Arndt, S; Okuda, J. Chem.
ReV. 2002, 102, 1953. (g) Cassani, M. C.; Gun’ko, Y. K.; Hitchcock,
P. B.; Hulkes, A. G.; Khvostov, A. V.; Lappert, M. F.; Protchenko,
A. V. J. Organomet. Chem. 2002, 647, 71. (h) Edelmann, F. T.;
Freckmann, D. M. M.; Schumann, H. Chem. ReV. 2002, 102, 1851.
(i) Evans, W. J. J. Organomet. Chem. 2002, 652, 61. (j) Izod, K. J.
Angew. Chem., Int. Ed. 2002, 41, 743.
(4) Hitchcock, P. B.; Holmes, S. A.; Lappert, M. F.; Tian, S. Chem.
Commun. 1994, 2691.
(5) Heckmann, G.; Niemeyer, M. J. Am. Chem. Soc. 2000, 122, 4227.
(6) Harder, S. Angew. Chem. 2004, 116, 2768; Angew. Chem., Int. Ed.
2004, 43, 2714.
(7) (a) Eaborn, C; Hitchcock, P. B.; Izod, K.; Lu, Z.-R.; Smith, J. D.
Organometallics 1996, 15, 4783. (b) Izod, K.; Clegg, W.; Liddle, S.
T. Organometallics 2000, 19, 3640. (c) Fedushkin, I. L.; Petrovskaya,
T. V.; Bochkarev, M. N.; Dechert, S.; Schumann, H. Angew. Chem.
2001, 113, 2540; Angew. Chem., Int. Ed. 2001, 40, 2474. (d) Weber,
F.; Sitzmann, H.; Schultz, M.; Sofield, C. D.; Andersen, R. A.
Organometallics 2002, 21, 3139.
(2) (a) Hong, S.; Marks, T. J. Acc. Chem. Res. 2004, 37, 673. (b) Hou,
Z.; Wakatsuki, Y. Coord. Chem. ReV. 2002, 231, 1.
(3) Gun’ko, Y. K.; Edelmann, F. T. Comments Inorg. Chem. 1997, 19,
153.
5320 Inorganic Chemistry, Vol. 46, No. 13, 2007
10.1021/ic700403d CCC: $37.00
© 2007 American Chemical Society
Published on Web 05/27/2007

Ytterbium(II) Hydride and Hydroxide Complexes
Scheme 1
Figure 1. Crystal structure of 3-Yb. All hydrogens, except the bridging
hydride ions, have been omitted for clarity.
hydride8 and hydroxide9 complexes have been reported. In
the course of our continuing interest in similarities and
differences in the chemistry of Ca2+ and Yb2+, we describe
here the synthesis and structure of an ytterbium(II) hydride
complex that is soluble in hydrocarbons and shows an
unusual high stability toward ligand exchange. Moreover,
we describe a synthetic route to the first Ln(II) hydroxide
complex, a class of compounds which were not accessible
yet. We specifically outline the differences with their calcium
analogues.
of the amide precursor 2-Ca with PhSiH3 gave good
conversion to 3-Ca. Therefore, following a similar route for
an analogue ytterbium complex seemed promising.
The heteroleptic precursor 2-Yb could be prepared by the
same convenient one-pot procedure that has been developed
for 2-Ca. Reaction of KN(SiMe3)2, YbI2, and DIPP-nacnacH
in a ratio of 2:1:1, respectively, gave 2-Yb as dark-violet
crystals in 54% yield. Subsequent reaction with PhSiH3 gave
the Yb(II) hydride complex 3-Yb in the form of dark-violet
crystals (27%).
Results and Discussion
Recently, the first hydrocarbon-soluble ytterbium(II) hy-
dride could be prepared by reaction of an alkyl precursor
with pressurized dihydrogen (eq 1).10 The extremely large
tridentate scorpionate ligand (TptBu, Me) in 1 provided suf-
ficient steric bulk and thus supressed ligand exchange (which
would result in formation of insoluble YbH2). In solution at
room-temperature complex 1 was stable for hours and
decomposed gradually into unidentified products.
The complex crystallized as a centrosymmetric dimer
(Figure 1) with symmetrically bridging hydride ions that
could be located and were isotropically refined as indepen-
dent atoms. The Yb-H bond lengths range from 2.21(4) to
2.23(4) Å and are within the standard deviation similar to
the average Yb-H bond of 2.26(3) Å in [(TptBu, Me)YbH]2
(1). The Yb···Yb distance (3.650(1) Å) in 1 is significantly
longer than that in 3-Yb (3.5204(2) Å), which suggests that
the steric bulk of the TptBu, Me ligand is larger than that of
the DIPP-nacnac/THF combination.
Yb(II) complexes generally crystallize isomorphous to
their Ca analogues. Complex 3-Ca crystallized from hexane
as a solvent-free crystal modification (Table 1, modification
A)8 and from benzene with a cocrystallized solvent mole-
cule (modification B). Structurally these modifications differ
in the orientation of the DIPP-nacnac ligands in respect to
each other. Complex 3-Yb, however, crystallized from
hexane in a completely different modification again
(Table 1). This might be due to the noticeable better solubility
of 3-Yb in hexane, which forced a crystallization at much
lower temperature and led to cocrystallization of a hexane
molecule.
Interestingly, attempts to synthesize a calcium hydride
complex with the very similar scorpionate ligand TptBu gave
formation of homoleptic (TptBu)2Ca in which one ligand is
bonded to Ca in (κ3-N,N′,N′′)-fashion and the other with two
pyrazole nitrogens and a B-H unit.8 Synthesis of a stable
calcium hydride complex 3-Ca was accomplished by use of
the sterically crowded DIPP-nacnac ligand system (Scheme
1; DIPP-nacnac ) CH{(CMe)(2,6-iPr2C6H3N)}2): reaction
(8) Harder, S.; Brettar, J. Angew. Chem. 2006, 118, 3554; Angew. Chem.,
Int. Ed. 2004, 45, 3474.
(9) Ruspic, C.; Nembenna, S.; Hofmeister, A.; Magull, J.; Harder, S.;
Roesky, H. W. J. Am. Chem. Soc. 2006, 128, 15000.
(10) Takats, J.; Ferrence, G. M.; McDonald, R. Angew. Chem. 1999, 111,
2372; Angew. Chem., Int. Ed. 1999, 38, 2233.
Inorganic Chemistry, Vol. 46, No. 13, 2007 5321

Ruspic et al.
Table 1. Cell Parameters and Selected Bond Distances (Å) for Dimeric Calcium and Ytterbium Hydride Complexes 3-Ca and 3-Yb
param
a (Å)
b (Å)
c (Å)
3-Ca (modification A)8
3-Ca (modification B)
3-Yb
13.2493(5)
14.5646(5)
18.9955(7)
90
93.654(2)
90
3658.1(2)
monoclinic
P21/n
2 half-dimers over i
1 n-hexane
2.21(4)-2.23(4)
2.22(4)
2.443(2)
2.376(2)-2.379(2)
47.980(2)
12.5912(4)
22.2426(8)
90
103.069(2)
90
13 089.2(8)
monoclinic
C2/c
19.7751(7)
19.7929(6)
18.2045(5)
90
99.231(2)
90
7033.1(4)
monoclinic
P21/c
2 half-dimers over i
2 half-C6H6 over i
2.16(4)-2.22(4)
2.18(4)
R (deg)
â (deg)
γ (deg)
V (Å3)
cryst system
space group
asym unit
cosolvent
M-H range
M-H av
M-O
1 dimer
2.09(4)-2.21(3)
2.15(3)
2.355(1)-2.391(1)
2.372(1)-2.390(1)
2.407(3)-2.415(3)
2.365(3)-2.378(3)
M-N
Comparison of the structure of 3-Yb with both 3-Ca
modifications shows that the ligand-metal bonding in these
compounds is similar (Table 1); bonds to Ca are slightly
shorter on account of its somewhat smaller ionic radius
(Ca2+, 1.00 Å; Yb2+, 1.02 Å).11 The best agreement is
observed between 3-Yb and 3-Ca (modification B), which
are both centrosymmetric dimers with a similar orientation
of the ligands.
The Yb(II)-hydride complex 3-Yb dissolves well in
benzene and is stable in solution at room temperature,
however, slowly decomposes at 75 °C into homoleptic
(DIPP-nacnac)2Yb and presumably YbH2. At 75 °C the half-
lifetime of 3-Yb in benzene is ca. 2.5 h. The analogue 3-Ca
is much more stable: whereas shortly heating a benzene
solution of 3-Ca to 75 °C did not result in noticeable
decomposition, prolonged heating at 75 °C gave decomposi-
tion into (DIPP-nacnac)2Ca and CaH2 with a half-lifetime
of ca. 24 h.
Recently, we observed unexpected close similarities in the
1H NMR spectra of benzylcalcium and benzylytterbium(II)
analogues.6 This is remarkable, since it is known that the
metal in benzyl complexes has a large influence on charge
delocalization and thus on the chemical shift of the ring
protons.12 1H NMR spectra of 3-Yb and 3-Ca (Figure 2) are
different: whereas the chemical shifts for most of the protons
compare well, a huge discrepancy is observed for the hydride
protons. The signal for the hydride hydrogen atom in 3-Ca
is a singlet at 4.45 ppm. The analogue resonance in 3-Yb is
found at 9.92 ppm, an unusually high chemical shift for a
proton with considerable hydride character. This value is
confirmed by two-dimensional NMR experiments and by the
clearly visible 171Yb satellites (JYb-H ) 398 Hz). The rather
high chemical shift as well as the 171Yb-1H coupling in 3-Yb
also compare well to the values obtained for [(TptBu,Me)YbH]2
(1): δ(H-) ) 10.50 ppm and JYb-H ) 369 Hz.
orbit-induced heavy-atom effects. These effects, which have
been well-described,13 are very pronounced for atoms close
to the heavy nuclei (r-3 dependence) and depend strongly
on the involvement of valence s-orbitals.
Recently, we reported on the catalytic hydrosilylation of
alkenes with early main-group metal complexes and proposed
catalytic cycles that explain the influence of polar solvents
on the regioselectivity of this reaction.14 Since a metal
hydride complex is considered the catalytical active species,
we tested the soluble complexes 3-Ca and 3-Yb in the
hydrosilylation of 1,1-diphenylethylene (DPE) with PhSiH3
(eq 2).
Both hydride complexes indeed were catalytically active
in this reaction, but the product distributions (Table 2) differ
substantially from each other and from that obtained with
the homoleptic catalyst: (R-Me3Si-ortho-Me2N-benzyl)2Ca·
2THF (9).14 The calcium catalysts 9 gave regioselectively
product 7 or 8, depending on the polarity of the medium.
Hydrosilylation with 3-Ca and 3-Yb, however, is indepen-
dent of the polarity of the solvent and gave in all cases 7 as
the major product. This difference is likely related to the
heteroleptic nature of the catalytically active species in
reactions with 3-Ca and 3-Yb.
The latter catalysts also resulted in formation of Ph2CHMe
as an unexpected byproduct, which is especially observed
for runs with 3-Ca under polar conditions. The mechanism
for formation of Ph2CHMe as well as the consequence of
metal choice on the product distributions are hitherto unclear
and encourage further investigation.
The ability of the DIPP-nacnac ligand in stabilizing
complexes with low coordination numbers and unique
functionalities allowed the synthesis of hydrocarbon-soluble
hydride complexes of Ca2+ and Yb2+. The unusual stability
of these complexes is presumably due to the rather strongly
bonded bidentate ligand and its bulky nature. This results in
The considerable differences in chemical shifts for the
hydride protons in 3-Ca and 3-Yb should not be explained
by differences in charge distributions in both complexes.
Rather we attribute this extreme downfield shift to spin-
(11) Shannon, R. D. Acta Crystallogr. 1976, A32, 751.
(12) (a) Hoffmann, D.; Bauer, W.; Hampel, F.; van Eikema Hommes, N.
J. R.; Schleyer, P. von R.; Otto, P.; Pieper, U.; Stalke, D.; Wright, D.
S.; Snaith, R. J. Am. Chem. Soc. 1994, 116, 528. (b) Feil, F.; Harder,
S. Organometallics 2000, 19, 5010.
(13) Kaupp, M.; Malkina, O. L.; Malkin, V. G.; Pyykko¨, P. Chem.sEur.
J. 1998, 4, 118.
(14) Buch, F.; Brettar, J.; Harder, S. Angew. Chem. 2006, 118, 2807; Angew.
Chem., Int. Ed. 2006, 45, 2741.
5322 Inorganic Chemistry, Vol. 46, No. 13, 2007

Ytterbium(II) Hydride and Hydroxide Complexes
Figure 2. 1H NMR spectra of the hydride complexes 3-Yb and 3-Ca in C6D6 at 20 °C.
Table 2. Product Distribution (%) in the Catalytic Hydrosilylation of
recently isolated in an Ar matrix at 6 K15 or could be detected
1,1-Diphenylethylene with Phenylsilanea
as a precipitate on the electrode in electrochemical reduction
of Ln(III) cations.16 The lack of a well-defined heteroleptic
Ln(II) hydroxide complex is inherent to the synthetic method
for hydroxide complexes. Whereas the first well-defined
group 3 hydroxide complex has been prepared by careful
hydrolysis of Cp3Y at -78 °C (eq 3),17 controlled hydrolysis
of Ln(II) complexes led to reduction of water and gave the
Ln(III) hydroxide complexes in >80% yields (eq 4).18
Interestingly, slow evaporation of water vapor into a THF
solution of decamethylsamarocene gave a highly sym-
metric hexanuclear Sm(III) cluster in which water is not
only oxidized to OH- but partially further reacted to O2-
(eq 5).19
catalyst
solvent
t (h)b
7
8
Ph2CHMe
9
9
16
3
5
4
4
100
0
79
53
99
95
0
0
0
19
42
1
THF
THF
THF
100
2
3-Ca
3-Ca
3-Yb
3-Yb
5
0
0
4
5
a Reactions at 50 °C; 2.5 mol % catalyst for 9 and 5 mol % catalyst for
3-Ca and 3-Yb. b Time indicating >98% conversion of one of the substrates
(1,1-diphenylmethane or phenylsilane).
Cp3Y(THF) H O8 Cp2Y-OH(THF) + CpH
(3)
(4)
2
Cp′2LnII H2O8 Cp′2LnIII-OH + 0.5H2
Cp′ ) Me3SiCp, Ln ) Yb; Cp′ ) 1,3-(Me3Si)2Cp, Ln ) Sm
Cp*2SmII H2O(g)8 (Cp*SmIII)6(O)3(OH)6
(5)
THF
The calcium hydroxide complex 4-Ca could be obtained
by two methods: (i) very slow diffusion of water vapor into
a solution of 2-Ca in hexane (the water phase was covered
by a hexane layer in order to slow down diffusion), in which
case it precipitated as well-defined crystals. (ii) By addition
of water to a THF solution of 2-Ca at -40 °C.9 Very slow
diffusion of water vapor into a solution of 2-Yb resulted in
Figure 3. Crystal structure of 4-Yb. All hydrogens, except those of the
hydroxide ions, have been omitted for clarity.
a stable cage around the central (MH)2 core and thus prevents
ligand exchange reactions. The recent isolation of an
isomorphous calcium hydroxide complex, [(DIPP-nacnac)-
CaOH·THF]2 (4-Ca, Scheme 1),9 is proof of principle that
this ligand system allows the isolation of calcium complexes
with other small anionic functionalities. The same ligand
environment could also enable the synthesis of the analogue
ytterbium(II) hydroxide complex.
(15) Xu, J.; Zhou, M. J. Phys. Chem. A 2006, 110, 10575.
(16) Steeman, E.; Temmerman, E.; Verbeek, F. J. Electroanal. Chem. 1978,
89, 113.
(17) Evans, W. J.; Hozbor, M. A.; Bott, S. G.; Robinson, G. H.; Atwood,
J. L. Inorg. Chem. 1988, 27, 1990.
(18) Hitchcock, P. B.; Lappert, M. F.; Prashar, S. J. Organomet. Chem.
1991, 413, 79.
(19) Evans, W. J.; Allen, N. T.; Greci, M. A.; Ziller, J. W. Organometallics
2001, 20, 2936.
Lanthanide(II) hydroxide compounds are not easily ac-
cessible: homoleptic Eu(OH)2 and Sm(OH)2 were only
Inorganic Chemistry, Vol. 46, No. 13, 2007 5323

Ruspic et al.
Table 3. Cell Parameter and Selected Bond Distances (Å) for Dimeric Calcium and Ytterbium Hydroxide Complexes 4-Ca and 4-Yb
complex
a (Å)
b (Å)
c (Å)
4-Ca (modification A)9
4-Ca (modification B)9
4-Yb
48.190(5)
12.595(1)
22.291(2)
90
102.850(6)
90
13191(2)
monoclinic
C2/c
12.6659(9)
12.7083(9)
13.2977(10)
98.423(6)
113.730(5)
110.982(6)
1720.6(2)
triclinic
P1h
1 half-dimer over i
1 half-toluene over i
2.221(2)-2.230(4)
2.226(3)
12.6955(15)
12.7651(15)
13.3877(14)
97.950(7)
114.063(7)
111.239(7)
1739.1(3)
triclinic
R (deg)
â (deg)
γ (deg)
V (Å3)
cryst system
space group
asym unit
cosolvent
M-OH range
M-OH av
M-O(THF)
M-N
P1h
1 dimer
1 half-dimer over i
1 half-toluene over i
2.264(6)-2.265(5)
2.265(6)
2.403(6)
2.433(5)-2.447(6)
2.209(2)-2.228(2)
2.219(2)
2.372(1)-2.449(1)
2.414(2)-2.426(1)
2.414(3)
2.419(3)-2.422(3)
a gradual discoloration of the intensely dark-purple solution
and gave rise to a color gradient and formation of a white
precipitate. We were not able to isolate well-defined products
from the reaction mixture, but the drastic change in color
and the paramagnetic behavior of products in NMR analysis
suggested oxidation to Yb(III) species. We presume water
acts in this reaction as oxidator and propose formation of
intermediates 5-Yb and 6-Yb which might form a variety
of products by ligand exchange (Scheme 1). The 0.32 V
(ii) 4-Yb crystalizes isomorphous to one of the two modi-
fications previously found for its Ca analogue (Table 3):
4-Ca and 4-Yb both crystallize from toluene in a triclinic
crystal system with similar cell parameters and one cen-
trosymmetric dimer and a toluene molecule in the unit cell.
Bond distances and angles are also strikingly similar (Table
3). (iii) The Yb-N bonds (average 2.445(6) Å) are much
longer than those in similar â-diketiminate complexes of Yb-
(III) (range: 2.27-2.33 Å).21 (iv) Complex 4-Yb was fully
characterized by two-dimensional 1H NMR analysis. Nearly
all chemical shifts are strikingly similar to those for 4-Ca.
Only the hydroxide group shows a resonance at -0.23 ppm
which is somewhat downfield to that in 4-Ca (-0.78 ppm)
and exhibits characteristic 171Yb satellites (2JYb-H ) 10.5 Hz).
A benzene solution of the Yb(II) hydroxide complex 4-Yb
is considerably more stable than the Yb(II) hydride complex
3-Yb: 2 h at 75 °C did not result in noticeable decomposi-
tion. Prolonged heating resulted in line-broadening of the
1H NMR signals. A possible decomposition route to Yb(III)
species could be further reduction of of 4-Yb according to
the reaction 4-Yb f 2 6-Yb + H2. Isolation of well-defined
reaction products from this decomposition reaction is in
progress.
difference in reduction potentials for the reactions Yb3+
+
e- f Yb2+ (-1.15 V) and H2O + e- f OH- + 0.5 H2
(-0.83 V) indicates that the conversion of 2-Yb to 5-Yb is
indeed thermodynamically allowed by ca. 15 kcal/mol.
Similar redox problems are also encountered in the syntheses
of Ln(II)-amine complexes, and Ln(III)-amide complexes
resulted instead: LnIIX2 + R2NH f (R2N)LnIIIX2 + 0.5 H2.20
Addition of a stoechiometric amount of water to a THF
solution of 2-Yb at -60 °C, however, did not result in
immediate discoloration, and we were able to crystallize an
intensely dark-green compound from a toluene solution at
low temperature. Although the intense color of the complex
1
is characteristic for an Yb(II) species, the H NMR signals
of the product dissolved in benzene-d6 were broadened
considerably, which is indicative for the presence of Yb-
(III). Purification by repeated recrystallization resulted in a
diamagnetic NMR sample.
A crystal structure determination (Figure 3) confirmed the
isolation of the ytterbium(II) hydroxide complex 4-Yb. The
compound crystallized as a centrosymmetric dimer in which
the OH- ions bridge the metal ions. The coordination
geometry around Yb is a distorted square pyramid in which
the oxygen atom of THF (O2) is at the apex and the nitrogen
and hydroxide ligands form the square base.
Although the hydrogen atoms of the hydroxide function-
alities could not be located unambiguously in this heavy-
atom structure, we have ample evidence that the current
complex indeed represents the first Ln(II) hydroxide and is
not an Yb(III) oxide, e.g., dimeric 6-Yb. (i) Its intense color
and diamagnetic behavior are typical for a Yb(II) species.
Conclusions
Hydrocarbon-soluble Yb(II) hydride complex 3-Yb is
easily accessible by reaction of an amide with phenylsilane
under mild conditions, a procedure which was also used in
the synthesis of the calcium analogue 3-Ca. Although the
crystal structures of 3-Yb and 3-Ca are not isomorphous,
their geometries are very similar. The 1H NMR spectrum of
3-Yb differs from that of 3-Ca: the signal for the hydride
hydrogen in 3-Yb is shifted 5.47 ppm downfield with respect
to that in 3-Ca, which is attributed to a spin-orbit-induced
heavy-atom effect. The heteroleptic Yb(II) hydride complex
is somewhat less stable to ligand exchange and formation
of insoluble metal dihydride than its calcium analogue. This
might be attributed to the slightly longer and weaker Yb-H
bond.
The first Ln(II) hydroxide complex 4-Yb could be obtained
by a controlled low-temperature hydrolysis of 2-Yb. Al-
(20) (a) Bochkarev, M. N.; Fagin, A. A. Chem.sEur. J. 1999, 5, 2990. (b)
Bochkarev, M. N.; Khoroshenkov, G. V.; Kuzyaev, D. M.; Fagin, A.
A.; Burin, M. E.; Fukin, G. K.; Baranov, E. V.; Maleev, A. A. Inorg.
Chim. Acta 2006, 359, 3315. (c) Warf, J. C. Angew. Chem. 1970, 82,
397; Angew. Chem., Int. Ed. Engl. 1970, 9, 383.
(21) Yao, Y.; Zhang, Z.; Peng, H.; Zhang, Y.; Shen, Q.; Lin, J. Inorg.
Chem. 2006, 45, 2175.
5324 Inorganic Chemistry, Vol. 46, No. 13, 2007

Ytterbium(II) Hydride and Hydroxide Complexes
temperature, stirred for further 30 min, and then dried in vacuo
(25 °C, 1 Torr, 30 min). The resulting green solid was washed with
hexane (2 × 5 mL) and again dried in vacuo (25 °C, 1 Torr, 30
min). Yield: 480 mg, 40%. Crystals suitable for X-ray structural
analysis could be obtained by slow cooling of a hot toluene solution
to -27 °C (the final yield is ca. 20%). Anal. Calcd for C33H50N2O2-
Yb (Mr ) 679.80): C, 58.30; H, 7.41. Found: C, 57.98; H, 7.57.
1H NMR (300 MHz, C6D6, 20 °C): δ ) -0.23 (s, 1H; Yb-OH),
though partial oxidation of Yb(II) is observed, the product
can be obtained Yb(III)-free by repeated crystallization.
Similar to its hydrocarbon-soluble calcium analogues, Yb-
(II) hydride and hydroxide complexes are well-soluble in
nonpolar solvents and could find possible application as a
molecular Yb(II) precursor in CVD or sol-gel syntheses of
Yb(II) or Yb(III) salts.
3
3
1.06 (d, J(H,H) ) 6.5 Hz, 12H; iPr), 1.21 (d, J(H,H) ) 6.5 Hz,
Experimental Section
3
12H; iPr), 1.23-1.60 (m, 10H; Me, THF), 3.19 (sept, J(H,H) )
6.5 Hz, 4H; iPr), 3.48 (m, 4H; THF), 4.70 (s, 1H; H-backbone),
7.02-7.12 (m, 6H; aryl). 13C NMR (75 MHz, C6D6, 20 °C): δ )
24.2 (iPr-Me), 24.5 (iPr-CH), 24.8 (iPr-Me), 25.2 (THF), 27.6
(backbone-Me), 68.4 (THF), 94.2 (backbone-CH), 123.3 (Ar), 123.4
(Ar), 141.3 (Ar), 146.8 (Ar), 162.7 (backbone-C). IR (nujol): ν˜
1509, 1462, 1403, 1377, 1313, 1261, 1225, 1168, 1098, 1017, 923,
880, 784, 722 cm-1. Mp: 198 °C (dec). Despite numerous attempts,
we do not observe a resonance for the O-H vibration. In some
cases we found a rather broad signal at 3423 cm-1 which we
attribute to water impurities rather than to the OH- stretching
frequency. IR spectra of 2-Ca showed a very weak but sharp
resonance at 3697 cm-1. A signal for the O-H stretching vibration
in the IR spectrum of matrix-isolated Sm(OH)2 was also absent.
This has been attributed to the weakness of such signals as predicted
by calculation.15
General Methods. Solvents were dried by standard methods and
distilled prior to use. All moisture- and air-sensitive reactions were
carried out under an inert argon atmosphere using standard Schlenk
techniques. Samples prepared for spectral measurements as well
as for reactions were manipulated in a glovebox. NMR spectra were
recorded on Bruker DPX300 and Bruker DRX500 spectrometers.
IR spectra were measured as nujol mull between KBr plates. Single
crystals have been measured on a Siemens SMART CCD diffrac-
tometer. Structures have been solved and refined using the programs
SHELXS-97 and SHELXL-97, respectively.22 All geometry cal-
culations and graphics have been performed with PLATON.23
Synthesis of 2-Yb. A 15 mL volume of THF was added to a
mixture of DIPP-nacnac-H (2.30 g, 5.49 mmol) and KN(SiMe3)2
(2.19 g, 10.98 mmol). Subsequently, a slurry of YbI2 (2.34 g, 5.49
mmol) in 10 mL of THF was added and the mixture was stirred
overnight. After centrifugation, the volatile components were
removed under vacuo (25 °C, 1 Torr, 30 min) and the resulting
purple solid was recrystallized by cooling a concentrated pentane
solution to -27 °C. The product crystallized in the form of large
purple blocks (2.50 g, 54%). Anal. Calcd for C39H67N3OSi2Yb (Mr
) 823.18): C, 56.90; H, 8.20. Found: C, 56.64; H, 8.38. 1H NMR
(300 MHz, C6D6, 20 °C): δ ) 0.20 (s, 18H; SiMe3), 0.86 (m, 4H;
General Procedure for the Catalytic Alkene Hydrosilylation.
A typical hydrosilylation experiment was carried out as follows:
A dry Schlenk-tube was charged with 1,1-diphenylethylene (2.0
mmol, dried by destillation from CaH2) and phenylsilane (2.0 mmol,
used as received). After addition of the catalyst (2.5 mol % for 9
and 5 mol % for 3-Ca and 3-Yb), the solution was heated to
50 °C. For experiments in THF ca. 1 mL of the solvent was added
before addition of the catalyst. The conversion was followed by
3
3
THF), 1.24 (d, J(H,H) ) 6.6 Hz, 12H; iPr), 1.36 (d, J(H,H) )
6.6 Hz, 12H; iPr), 1.62 (s, 6H; Me), 3.26-3.35 (m, 8H; iPr, THF),
4.77 (s, 1H; H-backbone), 7.10-7.15 (m, 6H; aryl). 13C NMR (75
MHz, C6D6, 20 °C): δ ) 5.8 (Me3Si), 24.7 (iPr-Me), 25.0 (THF),
25.4 (iPr-H), 25.5 (iPr-Me), 28.3 (backbone-Me), 69.3 (THF), 93.5
(backbone-CH), 124.0 (Ar), 124.6 (Ar), 141.3 (Ar), 147.0 (Ar),
165.4 (backbone-C). Mp: 150 °C (dec).
1
taking samples at regular time intervals and analysis by H NMR
and GC-MS. All hydrosilylation products and initiation products
have been isolated as pure compounds and have been completely
1
characterized by H, 13C, and 2D-NMR methods as well as by
GC-MS.
Crystal data for 3-Yb: measurement at -90 °C, Mo KR, 2θmax
) 57.2°, 9354 independent reflections (Rint ) 0.068), 7743
reflections observed with I > 2σ(I), monoclinic, space group P21/
n, cell parameters in Table 1, formula C66H100Yb2N4O2, Z ) 2, R
) 0.0282, wR2 ) 0.0705, GOF ) 1.08, Fmax ) +0.66 e Å-3, Fmin
) -0.73 e Å3. The bridging hydride atoms could be located and
have been refined. The rest of the hydrogen atoms were placed on
calculated positions and were refined in a riding mode. A
cocrystallized hexane molecule was found in the difference Fourier
but could not be refined satisfactorily due to disorder. The disorder
was treated with the SQUEEZE procedure incorporated in
PLATON.23
Synthesis of 3-Yb. Phenylsilane (129 mg, 1.19 mmol) was added
to a solution of 2-Yb (1.00 g, 1.19 mmol) in hexane (5.5 mL). The
solution was stirred for 1 h at 60 °C and then concentrated to half
of its original volume. Cooling to -27 °C gave 3-Yb in the form
of dark violett crystals (215 mg, 27%). Anal. Calcd for C33H50-
YbN2O (Mr ) 663.80): C, 59.71; H, 7.59. Found: C, 59.43; H,
7.70. 1H NMR (300 MHz, C6D6, 20 °C): δ ) 1.07 (d, 3J(H,H) )
3
6.9 Hz, 12H; iPr), 1.24 (d, J(H,H) ) 6.9 Hz, 12H; iPr), 1.41 (m,
3
4H; THF), 1.55 (s, 6H; Me), 3.17 (sept, J(H,H) ) 6.9 Hz, 4H;
iPr), 3.61 (m, 4H; THF), 4.68 (s, 1H, H-backbone), 7.05-7.15 (m,
6H, aryl), 9.92 ppm (s, satellites: 1J(171Yb,1H) ) 398 Hz, 1H, Yb-
H). 13C NMR (75 MHz, C6D6, 20 °C): δ )24.4 (iPr-Me), 24.9
(Me-backbone), 25.4 (THF), 25.9 (iPr-Me), 28.0 (iPr-CH), 69.5
(THF), 94.2 (backbone), 123.6 (Ar), 123.9 (Ar), 141.9 (Ar), 145.8
(Ar), 163.8 ppm (backbone). Mp: 185 °C (dec).
Synthesis of 4-Yb. Degassed water (32 µL, 1.78 mmol) was
added to a cooled (-60 °C) solution of 2-Yb (1.48 g, 1.79 mmol)
in 10 mL of THF. The mixture was allowed to warm to room
Crystal data for 4-Yb: measurement at -90 °C, Mo KR, 2θmax
) 53.0°, 8802 independent reflections (Rint ) 0.056), 8116
reflections observed with I > 2σ(I), triclinic, space group P1h, cell
parameters in Table 3, formula (C66H100Yb2N4O4)(C7H8), Z ) 1,
R ) 0.0475, wR2 ) 0.1455, GOF ) 1.11, Fmax ) +1.74 e Å-3
,
Fmin ) -1.74 e Å3. The crystal has been measured and re-
fined as nonmerohedral rotational twin with twin law
a
(1 0 0 0.85 1h 0 0.88 0 1h). The BASF value refined to 0.25. The
cocrystallized toluene molecule is placed over a center of in-
version and is consequently refined in a 50/50 disorder model. All
hydrogen atoms, including the hydroxide hydrogen atoms, have
been placed on calculated positions and were refined in a riding
mode.
(22) (a) Sheldrick, G. M. SHELXL-97, Program for Crystal Structure
Solution; Universita¨t Go¨ttingen: Go¨ttingen, Germany, 1997. (b)
Sheldrick, G. M. SHELXL-97, Program for Crystal Structure Refine-
ment; Universita¨t Go¨ttingen: Go¨ttingen, Germany, 1997.
(23) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool; Utrecht
University: Utrecht, The Netherlands, 2000.
Inorganic Chemistry, Vol. 46, No. 13, 2007 5325

Ruspic et al.
Crystallographic data (excluding structure factors) have been
deposited with the Cambridge Crystallographic Data Centre as
supplementary publication no. CCDC 638591 (3-Yb) or 631498
(4-Yb). Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge CB21EZ, U.K.
(fax: (+44)1223-336-033. e-mail: deposit@ccdc.cam.ac.uk).
Prof. Dr. R. Boese and Dr. M. Kirchner are kindly acknowl-
edged for collection of the X-ray data.
Supporting Information Available: Crystallographic data for
3-Yb and 4-Yb as CIF files. This material is available free of charge
via the Internet at http://pubs.acs.org.
Acknowledgment. We are grateful to the DFG for
supporting our research within the priority program SPP1166.
IC700403D
5326 Inorganic Chemistry, Vol. 46, No. 13, 2007