A comparative study on the reactivity of tris-β-diketiminate ytterbium complexes: Steric effect of β-diketiminato ligands

Article abstract of DOI:10.1002/ejic.201000156

A series of tris-β-diketiminate ytterbium complexes with the general formula [YbL₃] [L = [N(C₆H₅)C(Me)] ₂CH^- L^H (1); [N(4-MeC₆H ₄)C(Me)]₂CH^-, L

Full text of DOI:10.1002/ejic.201000156

FULL PAPER
DOI: 10.1002/ejic.201000156
A Comparative Study on the Reactivity of Tris-β-Diketiminate Ytterbium
Complexes: Steric Effect of β-Diketiminato Ligands
Rui Jiao,[a] Mingqiang Xue,[a] Xiaodong Shen,[a] Yong Zhang,[a] Yingming Yao,[a] and
Qi Shen*[a,b]
Keywords: N ligands / Polymerization / Ytterbium / Ligand effects / Steric hindrance
A series of tris-β-diketiminate ytterbium complexes with the
general formula [YbL3] {L = [N(C6H5)C(Me)]2CH–, LH (1);
[N(4-MeC6H4)C(Me)]2CH–, L4-Me (2), and [N(2-MeC6H4)-
C(Me)]2CH–, L2-Me (3)} were synthesized and structurally
characterized. All complexes have longer Yb–N bond lengths
than other YbL-containing derivatives, and complex 3 has
the longest average Yb–N bond length. A comparative study
on the reactivity of complexes 1–3 revealed that complex 3
was a highly active catalyst for the polymerization of ε-capro-
lactone and -lactide, as well as for the addition of amines to
L
carbodiimides, whereas both complexes 1 and 2 were almost
inactive under the same conditions. The active sequence is
consistent with the distance of the Yb–N bond, which is re-
flected in the sterically induced activation of the bulky tris-
β-diketiminate ytterbium complexes.
later lanthanide metals are not known. Considering the fact
that a normal Ln–β-diketiminate moiety in β-diketiminate
lanthanide dichlorides is inert for the ring-opening polymer-
ization of ε-CL, whereas the same moiety in sterically de-
manding tris-β-diketiminate complexes is highly active,[4]
the ring-opening polymerization of lactones including ε-CL
and -LA was chosen as a benchmark reaction to assess the
reactivity of Ln–β-diketiminate species. The activity of tris-
β-diketiminate ytterbium complexes towards the polymeri-
zation of ε-CL and -LA was compared in an attempt to
address the effect of the size of the β-diketiminato moiety
on the reactivity of Ln–β-diketiminate species. To explore
the potential application of these sterically demanding tris-
β-diketiminate lanthanide complexes in organic synthesis,
the addition of amines to carbodiimides was tested by using
these ytterbium complexes for the first time, as this addition
reaction is a direct approach to guanidines, which are im-
portant structural units in many biologically and pharma-
ceutically active compounds. It was found that the small
Introduction
Lanthanide derivatives bearing a β-diketiminato ancil-
lary ligand have been well documented,[1] but the synthesis
and chemistry of tris-β-diketiminate lanthanide complexes
have been quite limited. To the best of our knowledge, only
two kinds of complexes have been reported up to 2009: one
is a homoleptic tris-β-diketiminate complex bearing a β-di-
ketiminato ligand with a neighboring fused six-membered
heterocyclic ring[2] and the other is the same complex bear-
ing N,N-bisphenyl groups.[3] Recently, we studied the syn-
thesis and reactivity of tris-β-diketiminate lanthanide com-
plexes including [LnL4-Cl3] {L4-Cl = [N(4-ClC6H4)C(Me)]2-
CH–; Ln = Pr, Nd, Sm}, [NdLH3] {LH = [N(C6H5)C(Me)]2-
CH–}, and [NdL4-Me3] {L4-Me = [N(4-MeC6H4)C(Me)]2-
CH–} and found that these complexes can be synthesized
in high yields. They were also proven to serve as catalysts
for the ring-opening polymerization of ε-caprolactone (ε-
CL) and -lactide (-LA) with high activity.[4]
It is well known that the reactivity of sterically de-
manding [Ln(C5Me5)3] complexes is very sensitive to steric
factors, which can be tuned by changing the size of the
metal or the ligand.[5] To further explore the chemistry of
tris-β-diketiminate lanthanide complexes, the synthesis of
various tris-β-diketiminate complexes with a small Yb
metal was attempted; tris-β-diketiminate complexes with
metal ytterbium complexes [YbL3], where L was LH, L4-Me
,
and even more bulky L2-Me {L2-Me = [N(2-MeC6H4)C-
(Me)]2CH–} (Scheme 1), could be synthesized in good
yields and that their reactivity depended greatly on the size
[a] Key Laboratory of Organic Synthesis, College of Chemistry,
Chemical Engineering and Materials Science, Suzhou University,
Dushu Lake Campus, Suzhou 215123, P. R. China
Fax: +86-512-65880305
E-mail: qshen@suda.edu.cn
[b] State Key Laboratory of Organometallic Chemistry, Shanghai
Institute of Organic Chemistry, Chinese Academy of Sciences,
Shanghai 200032, P. R. China
Scheme 1.
Eur. J. Inorg. Chem. 2010, 2523–2529
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2523

R. Jiao, M. Xue, X. Shen, Y. Zhang, Y. Yao, Q. Shen
FULL PAPER
of the ligands. Whereas complexes with the less bulky li-
All the complexes are quite thermostable and decompose
gands LH and L4-Me are almost inactive for the ring-open- at 168–170, 150–153, and 175–176 °C for 1, 2, and 3,
ing polymerization of ε-CL or -LA and for the addition respectively, but they are all moderately sensitive to air and
of amines to carbodiimides, the more crowded complex moisture. They are freely soluble in donor solvents such as
[YbL2-Me3] is a highly active catalyst for these reactions un- thf and dimethoxyethane (dme) and moderately soluble in
der the same reaction conditions. Here we report the results. toluene.
The IR spectra of complexes 1, 2, and 3 exhibited the
strong absorptions near 1536, 1528, and 1512 cm–1, respec-
Results and Discussion
tively, which were consistent with partial C=N bond char-
Syntheses and Molecular Structures of [YbLH3] (1),
1
acter.[7] These complexes did not provide any resolvable H
NMR spectra; the resonances are broad and shifted due to
the strong paramagnetic nature of the Yb ion.
[YbL4-Me3] (2), and [YbL2-Me3] (3)
The reaction of anhydrous YbCl3 with LiLH in a 1:3 mo-
lar ratio in thf went smoothly to afford the homoleptic com-
plex [YbLH3] (1) as orange-yellow crystals in 75% yield
upon crystallization from toluene (Scheme 2).
Crystals of 1 and 2 suitable for X-ray diffraction determi-
nation were obtained by crystallization from toluene,
whereas crystals of 3 were obtained from a mixture of thf
and hexane. The molecular structures of complexes 1, 2,
and 3 are shown in Figures 1, 2, and 3 respectively. Selected
bond lengths and angles and details of the crystal data col-
lection are given in Tables 1 and 5.
Complex 3 crystallizes with one thf molecule in the unit
cell. The molecular structures of complexes 1–3 are similar.
The central Yb metal in each complex is ligated by six nitro-
gen atoms from three β-diketiminato groups in a distorted
octahedral geometry, which are the same as those for other
lanthanide metal analogues.[2–4] However, quite different
binding modes of Ln–β-diketiminate and bond parameters
among the three complexes were observed. Complexes 1
and 2 are isostructural, and each β-diketiminato ligand in
1 and 2 binds symmetrically to the central Yb atom. In
contrast, each β-diketiminato ligand in complex 3 coordi-
nates to the Yb atom in an asymmetrical binding mode.
The β-diketiminato skeletal atoms of each β-diketiminate in
1 and 2 are almost coplanar, whereas the five atoms of the
ligand L2-Me in complex 3 are not coplanar. All the Yb–N
bond lengths and all the N–Yb–N bond angles in the β-
diketiminato ligand are equal in complex 2 [2.360(5) Å for
the Yb–N bond and 79.85(18)° for the N–Yb–N angle] and
Scheme 2.
The same reaction with 3 equivalents of the lithium salt
LiL4-Me, after workup, yielded the corresponding complex
[YbL4-Me3] (2) as orange-yellow crystals in good yield
(Scheme 2). The structures of the two complexes were fur-
ther confirmed by X-ray structural analysis. The success in
the syntheses of complexes 1 and 2 spurred us to attempt
the synthesis of a ytterbium complex with the more bulky
ligand L2-Me to see whether the bulky complex [YbL2-Me
]
3
could be stabilized, as the deprotonation product was re-
ported when the synthesis of the ytterbium complex [Yb-
LiPr 3] {LiPr = [N(2,6-iPr2C6H3)C(Me)]2CH–} was tried by
an oxidation approach.[6] Thus, the reaction of YbCl3 with
the lithium salt LiL2-Me in a 1:3 molar ratio was conducted
at 50 °C in thf. Unfortunately, no definite complex could
be isolated. Then, the same reaction with the sodium salt
NaL2–Me was tried again in thf solution at 60 °C. We were
pleased to observe that the reaction proceeded perfectly, af-
ter workup, to generate a red product in good yield. The
product was further characterized by X-ray diffraction to
be the target complex [YbL2–Me3] (3; Scheme 3).
2
2
Figure 1. ORTEP diagram of 1 showing atom-numbering scheme;
thermal ellipsoids are drawn at the 10% probability level. Hydrogen
atoms are omitted for clarity.
Scheme 3.
2524
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2523–2529

Reactivity of Tris-β-Diketiminate Ytterbium Complexes
from 75.49(12) to 76.82(11)°, respectively. The average Yb–
N bond length of 2.398(3) Å in complex 3 is about 0.036 Å
longer than those for complexes 1 and 2 (2.363 Å in com-
plex 1 and 2.361 Å in complex 2). The average N–Yb–N
bond angle in complex 3 is 76.27(12)°, which is about 3°
smaller than those found in complexes 1 and 2. In compari-
son with the average Yb–N bond length [2.243(3)–
2.340(4) Å] for the trivalent LYb-containing complexes re-
ported,[8] the average Yb–N bond length [2.398(3) Å] in
complex 3 is 0.155–0.058 Å longer, whereas the average Yb–
N bond length in complexes 1 and 2 is 0.116–0.022 Å
longer. The average N–Yb–N angle in complexes 1, 2, and 3
is 79.23(10), 79.85(18), and 76.27(12)°, respectively, smaller
than 85.43(13)–81.25(15)° found in the YbL-containing
complexes mentioned above.[8] These bond parameters indi-
cate that the coordination environment around the central
metal Yb atom in tris-β-diketiminate ytterbium complexes
is more crowded than those in the trivalent YbL-containing
derivatives and complex 3, with the most bulky β-diketimin-
ato ligands, is the most sterically crowded.
Figure 2. ORTEP diagram of 2 showing atom-numbering scheme;
thermal ellipsoids are drawn at the 10% probability level. Hydrogen
atoms are omitted for clarity.
Comparative Reactivity of Complexes 1–3
The difference in the sterically crowded coordination en-
vironment around the Yb atom between complex 3 and
complex 1 or 2 might lead to the difference in reactivity of
Yb–β-diketiminate species. Thus, the reactivity of 1–3 for
the polymerization of lactones was examined.
Polymerization of ε-Caprolactone (ε-CL)
The activities of complexes 1 and 2 towards the ring-
opening polymerization of ε-CL (500 equiv. of ε-CL per
metal) were first assessed in a toluene solution at room tem-
perature. It was found that even after 24 h no polymeriza-
tion occurred for both systems, indicating that 1 and 2 were
completely inactive under the present conditions (Table 2,
Entries 6 and 8). However, a gradual increase in the vis-
cosity of the toluene solution containing ε-CL and 1 or 2
was observed when the polymerization temperature was
raised to 80 °C. For 1, a 73% yield of poly(ε-CL) was ob-
tained, and for 2, a 72% yield was obtained after 4 h
(Table 2, Entries 7 and 9). We have reported that the ana-
logues of early and middle lanthanide metals can serve as
highly active catalysts for this reaction at room tempera-
ture.[4] The observed large difference in activity demon-
strates that the size of the central metal has a remarkable
influence on the reactivity of these sterically demanding
complexes.
Figure 3. ORTEP diagram of 3 showing atom-numbering scheme;
thermal ellipsoids are drawn at the 10% probability level. Hydrogen
atoms are omitted for clarity.
Table 1. Selected bond lengths (Å) and bond angles (°) for com-
plexes 1–3.
1
3
2
Yb1–N1
Yb1–N2
Yb1–N3
Yb1–N4
Yb1–N5
Yb1–N6
(Yb–N)av.
N1–Ln1–N2 78.85(10)
N3–Ln1–N4 79.55(11)
N5–Ln1–N6 79.30(10)
(N–Ln–N)av. 79.23(10)
2.360(3)
2.360(3)
2.371(3)
2.355(3)
2.362(3)
2.368(3)
2.363(3)
2.442(3)
2.345(3)
2.403(3)
2.369(3)
2.438(3)
2.387(3)
2.398(3)
75.49(12)
76.50(12)
76.82(11)
76.27(12)
Yb1–N1
Yb1–N2
2.360(5)
2.361(5)
2.360(5)
2.361(5)
2.360(5)
2.361(5)
2.361(5)
79.85(18)
Yb1–N1A
Yb1–N2A
Yb1–N1B
Yb1–N2B
(Yb–N)av.
N1–Yb1–N2
N1B–Yb1–N2B 79.85(18)
N1A–Yb1–N2A 79.85(18)
(N–Ln–N)av.
In contrast, a rapid polymerization of ε-CL was observed
with the use of 3; poly(ε-CL) was obtained in 87% yield
after 15 min at room temperature with a molar ratio of 500
(monomer/catalyst; Table 2, Entry 1). When the molar ratio
of monomer/catalyst was increased to 1000, the monomer
79.85(18)
they are almost equal in complex 1 (2.355–2.371 Å and conversion still reached 97% within 60 min (Table 2, En-
78.85–79.55°); the bond parameters in complexes 1 and 2 try 4). Complex 3 is even more active than the reported β-
are well consistent with each other. However, the Yb–N diketiminate ytterbium aryloxo and amido chlorides
bond lengths and the N–Yb–N bond angles in complex 3 LiPr L Yb(OAr)Cl(thf), LMe Yb(OAr)Cl(thf) {LMe
=
1
2
2
2
are different and range from 2.345(3) to 2.442(2) Å and [N(2,6-Me2C6H3)C(Me)]2CH–}[9] and LiPr2YbNR2Cl.[8a]
Eur. J. Inorg. Chem. 2010, 2523–2529
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2525

R. Jiao, M. Xue, X. Shen, Y. Zhang, Y. Yao, Q. Shen
FULL PAPER
Table 2. Polymerization of ε-CL catalyzed by complexes 1–3.[a]
Entry
Initiator
[M]/[I]
Time (min)
T (°C)
Yield (%)[b] Mn,calcd ϫ 10–4[c] Mn,exp. ϫ 10–4[d]
PDI
1
2
3
4
5
6
7
8
9
3
3
3
3
3
1
1
2
2
500
1000
1000
1000
1500
500
500
500
500
15
15
30
60
15
180
240
180
240
25
25
25
25
25
25
80
25
80
87
43
85
97
46
–
73
–
72
4.97
4.91
9.70
11.1
7.88
–
4.97
–
4.97
5.41
5.11
13.5
21.7
34.0
–
5.21
–
5.72
1.88
1.58
1.61
1.54
1.61
–
1.57
–
1.61
[a] Polymerization conditions: in toluene; [ε-CL] = 0.82 . [b] Yield = weight of polymer obtained/weight of monomer used. [c] Mn,calcd.
= ([CL]/[I]) ϫ 114.14 ϫ (polymer yield). [d] Measured by GPC calibrated with standard polystyrene samples and corrected with the
coefficient of 0.56.
No end group was assigned to the β-diketiminato group,
but the signals for the ε-CL or -LA polymer were detected
by end-group analysis of ε-CL or -LA oligomers. The
same situation was also found in the cases of the tris-β-
diketiminate lanthanide complex as the catalyst.[4] Nor-
mally, no end-group detection was attributed to the forma-
tion of cyclic polymers through intramolecular attack of the
Ln–O bond in an active species to the N-bonded acyl car-
bon atom.[3b,12] Another possibility for no end-group detec-
tion may be because the molecular weights of the oligomers
A linear increase in the corrected molecular weights (Mn)
of the polymers with conversion and almost no changed
polydispersity indexes during the polymerization were ob-
served (Table 2, Entries 2–4). However, the values of Mn of
the resulting polyesters with complex 3 by 0.56[10] are much
larger than the calculated data for one polymer chain grow-
ing per metal center, and the polydispersity indexes of the
polymers (PDIs, evaluated by gel permeation chromatog-
raphy, GPC) are rather broad, ranging from 1.54 to 1.88.
This may be because of slow initiation and fast propaga-
tion, which result in low efficiency of the catalyst and rather
broad molecular weight distributions.
1
obtained are too large to be detected by H NMR spec-
troscopy.
Polymerization of L-Lactide (L-LA)
Addition of Amines to Carbodiimides
Complex 3 also showed very high -LA polymerization
To further explore the reactivity of tris-β-diketiminate yt-
activity. For instance, almost complete conversion of -LA terbium complexes, the addition of amines to carbodiimides
was obtained within 5 min at 80 °C in toluene with a mono- was examined by using complexes 1–3 for the first time.
mer/3 molar ratio of 300 (Table 3, Entry 1). When the molar
Taking the reaction of aniline with diisopropylcarbodi-
ratio of monomer/3 was increased to 500, the conversion of imide as a model reaction, the activities of complexes 1–3
the monomer was still as high as 96.2% after 5 min were tested at 60 °C under solvent-free conditions. As
(Table 3, Entry 2). The polymers obtained had corrected shown in Table 4, a great difference in the activities of 1, 2,
molecular weights[11] close to the calculated values and a and 3 was observed. The reaction with complex 3 as the
narrow polydispersity (1.18–1.30; Table 3, Entries 1–4). In catalyst afforded the product in almost quantitative yield,
contrast, both complexes 1 and 2 are inactive under the whereas the reaction with complex 1 or 2 yielded the prod-
same conditions (Table 3, Entries 5 and 6).
uct in a very low yield of 5 or 18%, respectively, at 1 mol-
Table 3. Polymerization of -LA by complexes 1–3.[a]
Entry
Initiator
[M]/[I]
Time (min)
T (°C)
Yield (%)[b] Mn,calcd ϫ 10–4[c] Mn,exp. ϫ 10–4[d]
PDI
1
2
3
4
5
6
3
3
3
3
2
1
300
500
700
800
300
300
5
5
30
60
300
300
80
80
80
80
80
80
98.3
96.2
54.0
29.6
–
4.25
6.93
5.45
3.41
–
5.39
7.63
6.94
4.66
–
1.30
1.21
1.26
1.18
–
–
–
–
–
[a] Polymerization conditions: in toluene; [-LA] = 1.0 . [b] Yield = weight of polymer obtained/weight of monomer used. [c] Mn,calcd.
=
([LA]/[I]) ϫ 144.13 ϫ (polymer yield). [d] Measured by GPC calibrated with standard polystyrene samples and corrected with the coeffi-
cient of 0.58.
2526
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2523–2529

Reactivity of Tris-β-Diketiminate Ytterbium Complexes
Table 4. Addition of amines to carbodiimides by complexes 1–3.[a] was observed. In complex 3, the three more bulky ligands
L2-Me around the Yb metal make the coordination sphere
Entry
Amine
Catalyst
Cat. loading Yield[b]
more crowded than those in complexes 1 and 2, as con-
firmed by structural analysis. Extreme steric saturation
could have limited the reactivity of 3. However, as a sub-
strate would not be able to approach the Yb metal, the
much longer length of the Yb–N bond observed in 3 in
comparison to that observed in complexes 1 and 2 (see the
Molecular Structure section) can provide a basis for the
high hemilability of the β-diketiminato ligand, as neither
the β-diketiminate nor the ytterbium receive the normal
electrostatic stabilization from each other at this distance.
Thus, a normally inert β-diketiminato ligand could be acti-
vated by “steric-induced activation”; that is, in an extremely
sterically crowded tris-β-diketiminate lanthanide complex
the metal–ligand bond can be elongated to such an extent
as to activate the ligand. Such an activation of ancillary
ligands was well documented in [Ln(C5Me5)3].[5]
If the activities of complexes 1 and 2 are compared to
those of larger metal analogues (Pr, Nd, Sm),[3b,4] it can be
found that for a given less bulky ligand LH or L4-Me the
complexes containing large metals (Pr, Nd) showed high
catalytic activity. This may be because the factor that ap-
pears to drive the reactivity of these complexes in the ring-
opening polymerization of lactones is the amount of coor-
dination unsaturation, not the amount of steric crowding.
In complexes 1 and 2 the small size of the Yb metal sur-
rounded by three of the ligands LH and/or L4-Me makes the
coordination sphere more crowded, which does not favor
the coordination of the substrate, whereas for large-metal
analogues there is still a room for coordination of the sub-
strate.
(mol-%)
(%)
1
2
3
4
5
6
7
8
C6H5NH2
C6H5NH2
C6H5NH2
3
3
1
2
3
3
3
3
3
3
3
3
1
0.25
1
1
1
1
1
1
1
96
88
5
18
94
NR
98
96
96
100
89
C6H5NH2
2-MeC6H4NH2
2,6-Me2C6H3NH2
4-FC6H4NH2
2-CH3OC6H4NH2
2-ClC6H4NH2
4-ClC6H4NH2
1-naphthylamine
morpholine
9
10
11
12
1
1
1
NR
[a] The reaction was performed by treating the amine (1 equiv.)
with the carbodiimide (1 equiv.) at 60 °C for 1 h. [b] Isolated yields;
NR = no reaction.
% catalyst loading (Table 4, Entries 1, 3, and 4). It is worth
noting that complex 3 is a highly active precatalyst for the
addition reaction. For example, the yield of the product still
reached 88% when the catalyst loading was decreased to
0.25 mol-% (Table 4, Entry 2). Preliminary results showed
that the reaction with aromatic amines bearing either elec-
tron-withdrawing or electron-donating groups, except for
the reaction performed with the bulky aromatic amine 2,6-
Me2C6H3NH2, went smoothly (Table 4, Entries 5–10) to
give the corresponding guanidines in excellent yields.
The catalytic reaction was supposed to proceed by the
active amide intermediate [L2YbNHR] formed in situ by
the reaction of YbL3 with amines directly, followed by in-
sertion of the carbodiimine to the amide and the proton-
ation of guanidinate by amine.[13] Thus, the formation of
the amide intermediate should be the key point for guan-
ylation. The amination of YbL3 with a bulky amine might
be not favored. This may be the reason why the reaction
with the bulky amine 2,6-Me2C6H3NH2 was sluggish. The
reaction of 3 with aniline was monitored by 1H NMR spec-
troscopy in [D6]benzene, and new signals for free HL2-Me
(at about 13.01, 2.19, and 1.78 ppm) appeared during the
period of the reaction, indicating the occurrence of the
amination of 3. Attempts to isolate the amide have not yet
been successful.
The results obtained from the reactions tested revealed
that the chemical behavior of complexes 1 and 2 is quite
similar: the ligands LH and L4-Me in 1 and 2 are both inert,
whereas complex 3 is highly active, and the normally inert
ligand L2-Me in 3 can act as an active group. The remarkable
difference in the reactivities of complexes 3 and 1 and/or 2
can only be attributed to the difference in amount of steric
crowding resulting from the different sizes of the ligands, as
Conclusions
In conclusion, we have synthesized and structurally char-
acterized the first tris-β-diketiminate ytterbium complexes
with various β-ketiminato ligands. Further examination of
their reactivities in the ring-opening polymerization of cap-
rolactone and lactide and in the addition of amines to car-
bodiimides revealed that the catalytic activities of the tris-
β-diketiminate ytterbium complexes were greatly affected
by the steric bulk of the β-diketiminato ligands. The most
sterically crowded complex 3 was found to be the most
active among the three complexes. Moreover, complex 3
was first explored to be a highly active precatalyst for the
addition of amines to carbodiimides. The results indicate
that a normally inert β-ketiminate ligand can become an
active group by steric-induced activation. Further study on
the chemistry of tris-β-diketiminate lanthanide complexes is
ongoing in our laboratory.
Experimental Section
General Procedures: All manipulations were performed under a
purified argon atmosphere by using standard Schlenk techniques.
no difference in the electronic factors of complexes 3 and 2 Solvents were degassed and distilled from sodium benzophenone
Eur. J. Inorg. Chem. 2010, 2523–2529
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2527

R. Jiao, M. Xue, X. Shen, Y. Zhang, Y. Yao, Q. Shen
FULL PAPER
ketyl before use. [D6]Benzene used for NMR reactions was dried
with Na and vacuum-transferred immediately prior to use. ε-Cap-
rolactone was purchased from Acros, dried with CaH2 for 48 h,
and distilled under reduced pressure. -Lactide was recrystallized
twice with dry toluene. LHH and L4-MeH were prepared according
to a literature method.[4] Anhydrous YbCl3 was prepared according
to a literature procedure.[14] Lanthanide analyses were performed
by EDTA titration with a xylenol orange indicator and a hexamine
buffer.[15] Carbon, hydrogen, and nitrogen analyses were performed
by direct combustion with a Carlo–Erba EA-1110 instrument. IR
spectra were recorded with a Nicolet-550 FTIR spectrometer as
C57H63N6Yb (1005.17): calcd. C 68.11, H 6.32, N 8.36, Yb 17.21;
found C 67.42, H 6.00, N 8.29, Yb 15.92. IR (KBr): ν = 1512 (w),
˜
1446 (w), 1369 (w), 1307 (w), 1229 (m), 1288 (m), 1099 (m), 1061
(m), 1022 (m), 965 (m), 887 (m), 779 (s), 659 (s), 459 (vs), 409 (vs)
cm–1.
Typical Procedure for the Polymerization of L-Lactide: A 50-mL
Schlenk flask, equipped with a magnetic stirring bar, was charged
with -lactide (0.50 g, 3.47 mmol) and toluene (3.47 mL). The con-
tents of the flask were then stirred at 80 °C until -lactide was dis-
solved, and then a toluene solution (2.00 mL) of complex 3
(11.24 mg, 0.0104 mmol, [LA]/[Yb] = 300:1, [LA] = 1.00 ) was
added by syringe. The mixture was stirred vigorously at 80 °C for
the desired time. The reaction mixture was quenched by ethanol
and precipitated in ethanol, filtered, washed with ethanol, and
dried in vacuo and weighed.
1
KBr pellets. H NMR spectra were obtained in CDCl3 for the li-
gands by using a Unity Inova-400 spectrometer. Melting points of
the crystalline samples were determined in sealed Ar-filled capillar-
ies. Molecular weights and molecular weight distributions were de-
termined against polystyrene standards by gel permeation
chromatography (GPC) at 30 °C with a Waters 1515 apparatus with
three HR columns (HR-1, HR-2, and HR-4) by using thf as the
eluent.
Typical Procedure for the Polymerization of ε-Caprolactone: A 50-
mL Schlenk flask, equipped with a magnetic stirring bar, was
charged with a solution of the initiator in toluene. To this solution
was added the desired amount of ε-caprolactone by syringe. The
contents of the flask were then stirred vigorously at the desired
temperature for a fixed time. The reaction mixture was quenched
by the addition of ethanol and then poured into ethanol to precipi-
tate the polymer. The polymer was dried in vacuo and weighed.
L2-MeH: A mixture of 2-toluidine (10.9 g, 0.1 mol), 2,4-pentane-
dione (5.1 g, 0.05 mol), and 4-toluenesulfonic acid (9.6 g) in toluene
(250 mL) was heated at reflux for 24 h in a Dean–Stark apparatus.
The toluene was then decanted off, and the white solid residue was
treated with diethyl ether (250 mL), water (100 mL), and Na2CO3
(12 g). After stirring for 30 min, the ether layer was separated, dried
with MgSO4, and evaporated in vacuo. The residue was dried in
vacuo (10–2 bar) at 100 °C for 6 h to remove any remaining free 2-
toluidine, then crystallized from hexane, and recrystallized from
anhydrous ethanol to give the ligand.[16] Yield: 12.1 g (87%). M.p.
35.9–36.2 °C. 1H NMR (400 MHz, CDCl3): δ = 12.54 (s, 1 H, NH),
7.16 (d, 4 H, ArH), 6.97 (d, 4 H, ArH), 4.91 (s, 1 H, β-CH), 2.20
(s, 6 H, o-CH3), 1.91 (s, 6 H, α-CH3) ppm.
Typical Procedure for Addition of Amines to Carbodiimides Cata-
lyzed by Complexes 1–3: Taking the reaction of aniline with di-
isopropylcarbodiimide as an example, A 30-mL Schlenk flask was
charged with complex 3 (0.0082 g, 0.0076 mmol). To the flask was
added N,NЈ-diisopropylcarbodiimide (0.12 mL, 0.76 mmol) and
aniline (0.07 mL, 0.76 mmol). The resulting mixture was stirred at
60 °C for a fixed time, then hydrolyzed with water (0.5 mL), ex-
tracted with dichloromethane (3ϫ10 mL), dried with anhydrous
Na2SO4, and filtered. After the solvent was removed under reduced
pressure, the residue was recrystallized from hexane to provide a
white solid N-phenyl-NЈ,N"-diisopropylguanidine. Yield: 0.1604 g
YbLH (1): To a slurry of anhydrous YbCl3 (1.12 g, 4.01 mmol) in
3
thf (25 mL) was slowly added a solution of LiLH (12.0 mmol) in thf
(12.0 mL) at room temperature. The reaction mixture was stirred at
50 °C for 48 h, the solvents were stripped off in vacuo, and toluene
was added to extract the product. The precipitate was removed by
centrifugation, and the orange-yellow supernatant was then con-
centrated and cooled to 0 °C to give orange-yellow crystals. Yield:
2.77 g (75%). M.p. 168–170 °C (dec.). C51H51N6Yb (921.02): calcd.
C 66.51, H 5.58, N 9.12, Yb 18.79; found C 66.99, H 5.53, N 9.07,
1
(96%). H NMR (CDCl3): δ = 7.22 (2 H), 6.95–6.91 (1 H), 6.86–
6.84 (2 H), 3.77 (2 H), 3.61 (2 H), 1.17–1.15 (12 H) ppm.
NMR-Scale Reaction: In a glove box, complex
3 (13.7 mg,
0.0127 mmol), C6D6 (0.5 mL), and aniline (25 mg, 0.2688 mmol)
were loaded into a J. Young NMR tube equipped with a Teflon
valve. The tube was closed and then removed from the glove box,
and the reaction was monitored by 1H NMR spectroscopy at room
temperature for the desired time.
Yb 18.82. IR (KBr): ν = 3056 (s), 2925 (m), 1628 (w), 1590 (m),
˜
1536 (vs), 1482 (s), 1451 (s), 1389 (vs), 1273 (s), 1189 (m), 1073 (w),
1027 (m), 934 (w), 818 (m), 749 (m), 702 (s), 648 (w), 509 (w) cm–1.
X-ray Crystallography: A suitable crystal was sealed in a thin-
walled glass capillary for X-ray structural analysis. Diffraction data
were collected with a Rigaku Mercury CCD area detector in the ω
scan mode by using graphite-monochromated Mo-Kα radiation (λ
= 0.71070 Å). The diffracted intensities were corrected for Lorentz
polarization effects and empirical absorption corrections. The
structures were solved by direct methods and expanded by Fourier
techniques. Atomic coordinates and thermal parameters were re-
fined by full-matrix least-squares procedures based on |F|2. All
non-hydrogen atoms were refined with anisotropic displacement
coefficients. Hydrogen atoms were treated as idealized contri-
butions. The structures were solved and refined using the
SHELXL-97 programs. Table 5 contains the crystallographic data
for complexes 1–3. CCDC-765303 (for 1), -765304
(for 2), and -765305 (for 3) contain the supplementary crystallo-
graphic 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.
YbL4-Me (2): Prepared in a manner similar to that used for the
3
preparation of 1, but YbCl3 (0.28 g, 1.00 mmol) and LiL4-Me
(15.0 mL, 3.00 mmol) were used instead. Orange-yellow crystals of
2 were obtained. Yield: 0.73 g (73%). M.p. 150–153 °C (dec.).
C57H63N6Yb (1005.17): calcd. C 68.11, H 6.32, N 8.36, Yb 17.21;
found C 68.73, H 6.15, N 8.01, Yb 17.21. IR (KBr): ν = 3025 (m),
˜
2925 (m), 2354 (m), 1996 (m), 1628 (s), 1528 (vs), 1505 (vs), 1447
(vs), 1273 (s), 1186 (s), 1027 (m), 857 (m), 741 (m) cm–1.
YbL2-Me (3): To a slurry of anhydrous YbCl3 (0.60 g, 2.14 mmol)
3
in thf (20 mL) was slowly added a solution of NaL2-Me (0.264  in
thf, 24.4 mL, 6.44 mmol) at room temperature. The reaction mix-
ture was stirred at 60 °C for 24 h. After the solvents were stripped
off in vacuo, the product was extracted with toluene (30 mL). The
toluene was then stripped off in vacuo, and hexane (20 mL) and
thf (1 mL) were added for crystallization at room temperature. Red
crystals were isolated. Yield: 1.14 g (53%). M.p. 175–176 °C (dec).
2528
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2523–2529

Reactivity of Tris-β-Diketiminate Ytterbium Complexes
Table 5. Crystallographic data for complexes 1–3.
1
2
3·thf
Empirical formula
Formula weight
Temperature (K)
Crystal system
Space group
Crystal size (mm)
a (Å)
b (Å)
c (Å)
α (°)
β (°)
C51H51N6Yb
921.02
193(2)
monoclinic
P 21/c
0.43ϫ0.17ϫ0.10
18.5059(15)
10.5351(7)
22.9125(19)
90
103.715(2)
90
4339.7(6)
4
1.410
2.197
C57H63N6Yb
1005.17
223(2)
C61H71N6OYb
1077.28
293(2)
trigonal
triclinic
¯
¯
P3
P1
0.36ϫ0.30ϫ0.12
19.022(2)
19.022(2)
23.977(2)
90
90
120
7513.1(14)
6
1.333
1.910
0.60ϫ0.50ϫ0.38
12.1778(14)
12.5956(15)
17.704(2)
95.185(3)
95.506(3)
96.273(3)
2672.8(5)
2
γ (°)
V (Å3)
Z
Dcalcd. (mgcm–3)
1.339
1.796
µ (mm–1)
F000
θ range (°)
Reflections collected/unique
Data/restraints/parameters
Goodness-of-fit on F2
Final R [IϾ2σ(I)]
wR2 (all data)
1876
3102
3.00–25.50
19321/9279 (Rint = 0.0413)
9279/0/589
1.156
0.0626
0.1182
1114
3.06–27.48
47470/9948 (Rint = 0.0458)
9948/0/530
1.198
0.0440
0.0704
3.02–25.35
25863/9700 (Rint = 0.0317)
9700/11/621
1.109
0.0382
0.0947
[4]
[5]
[6]
[7]
[8]
M. Q. Xue, R. Jiao, Y. Zhang, Y. M. Yao, Q. Shen, Eur. J.
Inorg. Chem. 2009, 4110–4118.
a) W. J. Evans, B. L. Davis, Chem. Rev. 2002, 102, 2119–2136;
b) W. J. Evans, Coord. Chem. Rev. 2000, 206–207, 263–283.
P. B. Hitchcock, M. F. Lappert, A. V. Protchenko, Chem. Com-
mun. 2005, 951–953.
D. S. Richeson, J. F. Mitchell, K. H. Theopold, J. Am. Chem.
Soc. 1987, 109, 5868–5870.
a) Y. M. Yao, Z. Q. Zhang, H. M. Peng, Y. Zhang, Q. Shen, J.
Lin, Inorg. Chem. 2006, 45, 2175–2183; b) Y. M. Yao, Y. Zhang,
Q. Shen, K. B. Yu, Organometallics 2002, 21, 819–824; c) Y. M.
Yao, M. Q. Xue, Y. J. Luo, Z. Q. Zhang, R. Jiao, Y. Zhang, Q.
Shen, W. T. Wong, K. B. Yu, J. Sun, J. Organomet. Chem. 2003,
678, 108–116.
Acknowledgments
We are grateful to the National Natural Science Foundation of
China (Grant no. 20632040) for financial support.
[1] a) L. Bourget-Merle, M. F. Lappert, J. R. Severn, Chem. Rev.
2002, 102, 3031–3065; b) P. G. Hayes, W. E. Piers, R. McDon-
ald, J. Am. Chem. Soc. 2002, 124, 2132–2133; c) P. G. Hayes,
W. E. Piers, M. Parvez, J. Am. Chem. Soc. 2003, 125, 5622–
5623; d) D. Neculai, H. W. Roesky, A. M. Neculai, J. Magull,
R. Herbst-Irmer, Organometallics 2003, 22, 2279–2283; e) G. B.
Nikiforov, H. W. Roesky, T. Labahn, D. Vidovic, D. Neculai,
Eur. J. Inorg. Chem. 2003, 433–436; f) A. G. Avent, P. B. Hitch-
cock, A. V. Khvostov, M. F. Lappert, A. V. Protchenko, Dalton
Trans. 2004, 2272–2280; g) Z. Q. Zhang, Y. M. Yao, Y. Zhang,
Q. Shen, W. T. Wong, Inorg. Chim. Acta 2004, 357, 3173–3180;
h) L. K. Knight, W. E. Piers, P. Fleurat-Lessard, M. Parvez, R.
McDonald, Organometallics 2004, 23, 2087–2094; i) M. Q.
Xue, Y. M. Yao, Q. Shen, Y. Zhang, J. Organomet. Chem. 2005,
690, 4685–4691; j) P. G. Hayes, W. E. Piers, M. Parvez, Organo-
metallics 2005, 24,1173–1183; k) C. Cui, A. Shafir, J. A. R.
Schmidt, A. G. Oliver, J. Arnold, Dalton Trans. 2005, 1387–
1393; l) D. V. Vitanova, F. Hampel, K. C. Hultzsch, Dalton
Trans. 2005, 1565–1566; m) L. K. Knight, W. E. Piers, R. Mc-
Donald, Organometallics 2006, 25, 3289–3292; n) E. Lu, W.
Gan, Y. Chen, Organometallics 2009, 28, 2318–2324; o) A. L.
Kenward, W. E. Piers, M. Parvez, Organometallics 2009, 28,
3012–3020; p) A. L. Kenward, J. A. Ross, W. E. Piers, M. Par-
vez, Organometallics 2009, 28, 3625–3628.
[9]
M. Q. Xue, H. M. Sun, H. Zhou, Y. M. Yao, Q. Shen, Y.
Zhang, J. Polym. Sci. Polym. Chem. Ed. 2006, 44, 1147–1152.
M. Save, M. Schappacher, A. Soum, Macromol. Chem. Phys.
2002, 203, 889–899.
[10]
[11] A. Kowalski, A. Duda, S. Penczek, Macromolecules 1998, 31,
2114–2122.
[12] a) K. C. Hultzsch, T. P. Spaniol, J. Okuda, Organometallics
1997, 16, 4845–4856; b) J. L. Chen, Y. M. Yao, Y. J. Luo, L. Y.
Zhou, Y. Zhang, Q. Shen, J. Organomet. Chem. 2004, 689,
1019–1024.
[13] W. X. Zhang, M. Nishiura, Z. Hou, Chem. Eur. J. 2007, 13,
4037–4051.
[14] M. D. Taylor, C. P. Carter, J. Inorg. Nucl. Chem. 1962, 24, 387–
391.
[15] J. L. Atwood, W. E. Hunter, A. L. Wayda, W. J. Evans, Inorg.
Chem. 1981, 20, 4115–4119.
[16] P. H. M. Budzelaar, A. B. Oorta, A. G. Orpenb, Eur. J. Inorg.
Chem. 1998, 1485–1494.
[2] A. Mandel, J. Magull, Z. Anorg. Allg. Chem. 1995, 621, 941–
944.
[3] a) D. Drees, J. Magull, Z. Anorg. Allg. Chem. 1994, 620, 814–
818; b) D. Barbier-Baudry, A. Bouazza, C. H. Brachais, A.
Dormond, M. Visseaux, Macromol. Rapid Commun. 2000, 21,
213–217.
Received: February 8, 2010
Published Online: May 7, 2010
Eur. J. Inorg. Chem. 2010, 2523–2529
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2529