The selective insertion of carbon dioxide into a lanthanide(III) 2,6-di-t-butyl-phenoxide bond

Article abstract of DOI:10.1016/j.poly.2012.05.021

An investigation of the CO₂(g) insertion products for a series of fully characterized monomeric lanthanide 2,6-di-t-butyl-phenoxide compounds ([Ln(DBP)₃]; Ln = Ce (1), Sm (2), Dy (3), Y (4), Er (5), Yb (6), and Lu (7)) was undertaken at low pressure (<5 psi). From the products isolated, only one CO₂(g) molecule per molecule [Ln(DBP)₃] was found to insert, forming either the [Ce(μ_c-O₂C-DBP)(DBP) ₂]₂ (8) or [Ln(μ-O₂C-DBP)(DBP) ₂]₂ (Ln = Sm (9), Dy (10), Y (11), Er (12), Yb (13), and Lu (14)) structure. The purity of the bulk powders of 8-14 were verified by FT-IR and elemental analyses; however solution structures could not be studied due to the low solubility of the complexes. Higher pressures to increase the degree of CO₂(g) insertions did not alter the degree of substitution. The selectivity of CO₂(g) insertion was attributed to an interaction of the methyl moieties of the DBP ligand blocking coordination sites on the Ln metal center, as observed in the solid and solution state of 1-7.

Full text of DOI:10.1016/j.poly.2012.05.021

Polyhedron 42 (2012) 258–264
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
The selective insertion of carbon dioxide into a lanthanide(III)
2,6-di-t-butyl-phenoxide bond
a,b,
c
Leigh Anna M. Steele a, Timothy J. Boyle a, , Richard A. Kemp
, Curtis Moore
⇑
⇑
a Sandia National Laboratories, Advanced Materials Laboratory, 1001 University Boulevard, SE, Albuquerque, NM 87106, United States
b University of New Mexico, Department of Chemistry and Chemical Biology, 1 University of New Mexico, Albuquerque, NM 87131, United States
c University of California, San Diego, Department of Chemistry and Biochemistry, 9500 Gilman Drive, La Jolla, CA 92093-0358, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
An investigation of the CO2(g) insertion products for a series of fully characterized monomeric lanthanide
2,6-di-t-butyl-phenoxide compounds ([Ln(DBP)3]; Ln = Ce (1), Sm (2), Dy (3), Y (4), Er (5), Yb (6), and Lu
(7)) was undertaken at low pressure (<5 psi). From the products isolated, only one CO2(g) molecule per
Received 20 February 2012
Accepted 21 May 2012
Available online 28 May 2012
molecule [Ln(DBP)3] was found to insert, forming either the [Ce(lc-O2C-DBP)(DBP)2]2 (8) or [Ln(l-O2C-
DBP)(DBP)2]2 (Ln = Sm (9), Dy (10), Y (11), Er (12), Yb (13), and Lu (14)) structure. The purity of the bulk
powders of 8–14 were verified by FT-IR and elemental analyses; however solution structures could not be
studied due to the low solubility of the complexes. Higher pressures to increase the degree of CO2(g)
insertions did not alter the degree of substitution. The selectivity of CO2(g) insertion was attributed to
an interaction of the methyl moieties of the DBP ligand blocking coordination sites on the Ln metal cen-
ter, as observed in the solid and solution state of 1–7.
Keywords:
CO2(g) insertion
Lanthanide alkoxides
Carbonates
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
hetero-atoms of the –OR ligands result in irregular geometric
arrangements and a wide range of coordination environments
Lanthanide alkoxides ([Ln(OR)3]) have become of interest for
the production of ceramic materials [1] due to a number of inher-
ent properties this family of compounds possesses. In particular,
the lanthanide contraction [2] imparts a smooth and systematic
change in cation size without altering the charge. This allows for
fine-tuning of the properties of materials through subtle structural
and electronic changes as different Ln cations are substituted. As
an example, the fatigue behavior of the perovskite phase of lead
zirconium titanate (PZT) materials was successfully decreased
through the doping of PZT with the aliovalent Dy cation as deter-
mined from the systematic study of Ln-doped PZT with a series
of [Ln(OR)3] [3]. Further, the inherent luminescent properties of
several of the Ln cations allow for their use in such diverse research
arenas as bio-imaging agents (i.e., upconvert materials) that mimic
naturally occurring fluorescent minerals [1a] or scintillator materi-
als [4] for detecting radioactive materials.
[6]. This phenomenon, coupled with the paramagnetic nature of
a number of these compounds, typically requires crystal structure
investigations in order to fully characterize the [Ln(OR)3] precur-
sors. Knowing the structural properties is helpful in optimizing
the processing parameters for the final ceramic materials of inter-
est. Often, the as-generated oxide material is not crystalline and re-
quires thermal treatment to obtain the desired ceramic oxide
phase. Unfortunately, as the [Ln(OR)3] complexes are thermally
treated, they often form carbonates through absorption of carbon
dioxide [CO2(g)] from circumjacent atmosphere, ligand decompo-
sition, or both. The resultant carbonates require much higher pro-
cessing temperatures to convert them into the desired oxide phase,
an undesirable attribute that has limited their utility. Understand-
ing this CO2(g) insertion process and how to control it is of interest
to reduce the thermal budget of final materials produced.
While research concerning the insertion of small molecules such
as CO2(g) into the metal-oxygen bonds of transition metal [M(OR)n]
complexes is voluminous, few studies concerning the reactivity of
CO2(g) with [Ln(OR)3] compounds are available [7]. Furthermore,
no structurally characterized lanthanide carbonate [Ln(O2COR)3]
compounds that are derived from the insertion of CO2(g) into Ln–
OR bonds have been reported [6b,8]. Comparative studies with
transition metals would not benefit in understanding this reactivity
due to the covalent nature of the transition metal [M(OR)n] species
versus the more ionic nature of the [Ln(OR)3] bonding [5].
The predominantly ionic bonding character that exists for
[Ln(OR)3] is due to the ‘buried’ f-orbitals that reside in lower en-
ergy levels than the outer d-orbitals [5]. Thus, the primarily elec-
trostatic interactions between the large Ln cations and the
⇑
Corresponding authors. Tel.: +1 505 272 7625; fax: +1 505 272 7336 (T.J. Boyle),
tel.: +1 505 272 7609; fax: +1 505 272 7336 (R.A. Kemp).
E-mail addresses: tjboyle@Sandia.gov (T.J. Boyle), rakemp@UNM.edu (R.A. Kemp).
0277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.05.021

Leigh Anna M. Steele et al. / Polyhedron 42 (2012) 258–264
259
The focus of this research was to determine the extent and
structural changes that occur from the insertion of CO2(g) into
the Ln–O bonds of [Ln(OR)3] compounds as a means to understand
the thermal conversion of these precursors into ceramic materials.
To achieve this goal, a series of structurally similar [Ln(OR)3] pre-
cursors that were coordinatively unsaturated was necessary to
serve as the precursor compounds. The alcohol 2,6-di-t-butylphe-
nol (2,6-[(CH3)3C]2C6H3OH or H-DBP ligand) was selected for use,
mainly due to its steric bulk and the previously known, tri-coordi-
nated monomeric structures reported for [Ln(DBP)3] (Ln = Ce [9], Pr
[10], Nd [10], and Dy [3]). Since not all of the [Ln(DBP)3] complexes
are known as structurally characterized materials, we were inter-
ested in expanding and preparing a series of monomeric and unsol-
vated complexes. This was successfully achieved where Ln = Ce (1)
[9], Sm (2), Dy (3) [3], Y (4), Er (5) Yb (6), and Lu (7) compounds.
These cations were chosen to represent a range of the lighter and
heavier Ln cations. Once characterized, the monomeric compounds
Anal. Calc. C42H63O3Sm (MW = 766.36 g/mol): C, 65.83; H, 8.29. Found:
C, 66.16; H, 8.34%.
[Y(DBP)3] (4): Used [Y(N(SiMe3)2)3] (0.50 g, 0.88 mmol), H-DBP
(0.54 g, 2.6 mmol) and ꢀ10 mL of tol. No color change was ob-
served. Yield 0.23 g (37%). FT-IR (KBr, cmꢁ1): 3650(m), 2956(s),
2917(s,sh), 2872(w,sh), 2368(w), 2345(w), 1459(w), 1426(w,sh),
1407(s), 1384(m,sh), 1360(m,sh), 1316(m), 1240(s), 1199(m),
1123(w), 1103(m), 998(m), 925(w), 865(s), 821(s), 806(s),
795(w,sh), 751(s), 695(m), 658(s), 588(w,sh), 548(s), 527(w,sh),
450(s). 13C{1H} NMR (100.5 MHz, C7D8) d 154.2, 136.0, 120.3
(C6H3), 34.3 [–C(CH3)3], 30.3 [–C(CH3)3]. Anal. Calc. C42H63O3Y
(MW = 704.83 g/mol): C, 71.55; H, 9.01. Found: C, 72.15; H, 9.46%.
[Er(DBP)3] (5): Used [Er(N(SiMe3)2)3] (0.50 g, 0.77 mmol), H-
DBP (0.48 g, 2.3 mmol) and ꢀ10 mL of tol. No color change was ob-
served. Yield 0.52 g (87%). FT-IR (KBr, cmꢁ1): 3651(s), 3081(w, sh),
2959(s), 2872(w,sh), 2369(w), 1412(s), 1387(s), 1358(m),
1315(w,m) 1260(s), 1196(m), 1098(s), 1022(m,sh), 976(m,sh),
936(w), 875(s), 820(m), 807(m), 748(s), 663(s), 607(m), 548(s),
453(s). Anal. Calc. C42H63O3Er (MW = 782.74 g/mol): C, 64.41; H,
8.11. Found: C, 64.32; H, 8.99%.
1–7 were treated with CO2(g) to form [Ce(
lc-O2C-DBP)(DBP)2]2 (8)
(‘lc denotes chelating bridging) and [Ln(
’
l
-O2C-DBP)(DBP)2]2
[Ln = Sm (9), Dy (10), Y (11), Er (12), Yb (13), and Lu (14)]. The syn-
theses and characterization of 1–14 will be discussed in detail.
[Yb(DBP)3] (6): Used [Yb(N(SiMe3)2)3] (0.50 g, 0.77 mmol), H-
DBP (0.55 g, 2.7 mmol) and ꢀ10 mL of tol. Color changed from
red to dark red/brown. Yield 0.54 g (89%). FT-IR (KBr, cmꢁ1):
3648(m), 2951(m), 2362(w), 1457(s), 1246(s), 1361(m), 1315(m),
1250(w), 1231(m, sh), 1198(m), 1143(s), 1122(m), 1023(w),
868(m), 844(m), 821(s, sh), 806(m), 795(m), 747(s), 667(m),
588(m), 446(m), 419(s). 13C{1H} NMR (100.5 MHz, C7D8) d 154.3,
135.8, 120.3 (C6H3), 34.3 [–C(CH3)3], 30.4 [–C(CH3)3]. Anal. Calc.
2. Materials and methods
All compounds described below were handled with rigorous
exclusion of air and water using standard Schlenk line and glove
box techniques unless otherwise noted. The following reagents
and solvents were used as received (Aldrich): H-DBP, LnCl3 (Ln = Ce,
Sm, Dy, Y, Er, Yb, Lu), KN(SiMe3)2, THF, and toluene. CO2(g) in
99.999% purity was supplied by Trigas (8200 Washington Street
Northeast, Albuquerque, NM 87113). [Ln(NR2)3] were synthesized
from the reaction of anhydrous LnCl3 and KN(SiMe3)2 and the prod-
ucts were characterized and found to be in agreement with the lit-
erature [11].
All analyses were performed on dry crystalline materials. FT-IR
spectroscopic data were obtained using KBr pressed pellets using a
Bruker Vector 22 Instrument under an atmosphere of flowing
nitrogen. Elemental analyses were collected on a Perkin-Elmer
2400 CHN–S/O elemental analyzer. Solution state 13C{1H} NMR
spectra were obtained using a 5 mm BB solution probe on a Bruker
AMX 400 MHz spectrometer. All spectra were referenced against
the protons in deuterated toluene-d8. Samples were made up in a
glovebox using crystalline material and the sample tubes were
flame sealed under vacuum to avoid reaction with the atmosphere.
C42H63O3Yb (MW = 789.00 g/mol): C, 63.94; H, 8.05. Found: C,
63.76; H, 8.52%.
[Lu(DBP)3] (7): Used [Lu(N(SiMe3)2)3] (0.50 g, 0.76 mmol), H-
DBP (0.55 g, 2.7 mmol) and ꢀ10 mL of tol. Changed from pale yel-
low to light purple. Yield 0.39 g (65%). FT-IR (KBr, cmꢁ1): 3645(m),
3081(w, sh), 2957(m), 1457(m), 1425(m), 1388(s, sh), 1360(s, sh),
1316(s, sh), 1250(m, sh), 1231(s), 1196(s, sh), 1143(s), 1122(s),
1095(s, sh), 1023(m), 878(s), 844(m), 821(m), 806(s, sh), 795(s),
746(s), 667(m), 588(m), 527(m), 449(m), 419(m). 13C{1H} NMR
(100.5 MHz, C7D8) d 154.2, 135.8, 120.3 (C6H3), 35.0 [–C(CH3)3],
30.4 [–C(CH3)3]. Anal. Calc. C42H63LuO3 (MW = 790.89 g/mol): C,
63.78; H, 8.03. Found: C, 64.21; H, 7.98%.
2.2. CO2(g) insertion synthesis
A 53 mM solution of the appropriate precursor (1–7) dissolved
in toluene was put into a two-neck round bottom reaction flask
equipped with a nitrogen adaptor and a rubber septum. The flask
containing the solution was carefully transferred from the glove-
box to the Schlenk line and placed under an Ar atmosphere. An
oil bubbler was used to ensure a positive pressure of gas was main-
tained in the flask. Technical grade CO2(g) (5 psig) was purged
through tubing equipped with a syringe needle for 2 min to clean
the line of air and yield a positive pressure of CO2(g). The needle
was then stuck into the septum and the argon shut off so that
the oil bubbler could monitor CO2(g) flow. The solution was stirred
under an atmosphere of CO2(g) for 30 min during which no color
changed was observed. Some of the volatile components were re-
moved via vacuum distillation and the remaining solution was
transferred to the glovebox for slow evaporation crystal growth.
2.1. [Ln(DBP)3] syntheses
A general synthesis will be described, since the preparation of
all the [Ln(DBP)3] compounds were identical. To a stirring solution
of [Ln(N(SiMe3)2)3] dissolved in ꢀ5 mL of toluene under an argon
atmosphere, a solution of three equivalents of H-DBP in toluene
(ꢀ5 mL) was added drop-wise. For most of the reactions a color
change occurred immediately and the reaction was allowed to stir
for 12 h. After this time the reaction mixture was concentrated by
slow evaporation of the volatile components until X-ray quality
crystals were formed. Compounds 1 [9] and 3 [3] have been pub-
lished previously.
[Sm(DBP)3] (2): Used [Sm(N(SiMe3)2)3] (0.50 g, 0.79 mmol), H-DBP
(0.57 g, 2.8 mmol) and ꢀ10 mL of tol. Color changed from pale yellow
to orange. Yield 0.50 g (82%). FT-IR (KBr, cmꢁ1): 3642(m), 3079(w),
2957(m), 1583(s), 1411(s), 1386(m, sh), 1358(s, sh), 1347(s),
1244(m), 1195(s), 1143(s), 1124(s), 1098(m), 869(s), 819(s), 796(s,
sh), 748(s), 658(s), 564(s), 548(s), 452(s). 13C{1H} NMR (100.5 MHz,
C7D8) d 154.2, 135.9, 120.3 (C6H3), 34.3 [–C(CH3)3], 30.3 [–C(CH3)3].
[Ce(DBP)2(lc-O2C-DBP)]2 (8): Used 1 (0.60 g, 0.79 mmol) in
ꢀ15 mL toluene with no color change noted. Yield 0.20 g (32%).
FT-IR (KBr, cmꢁ1) 3643(m), 2960(s), 2873(w, sh), 1625(w, sh),
1589(s), 1577(m, sh), 1482(w), 1426(s), 1411(w, sh), 1369(s),
1349(s), 1232(m, sh), 1248(s), 1195(m), 1176(m), 1143(s),
1105(s), 1039(w, sh), 1024(m), 861(s), 844(s, sh), 819(s), 807(s),
796(s), 747(s), 722(s), 676(m), 653(s), 588(m), 544(m), 450(s).

260
Leigh Anna M. Steele et al. / Polyhedron 42 (2012) 258–264
Anal. Calc. for C86H126Ce2O10 (MW = 1600.11 g/mol): C, 64.55; H,
7.94. Found: C, 64.10; H, 8.04%.
heavy atoms. Subsequent Fourier syntheses yielded the remaining
light-atom positions. The hydrogen atoms were fixed in positions
of ideal geometry and refined using SHELX software [12a]. The final
refinement of each compound included anisotropic thermal
parameters for all non-hydrogen atoms. Table 1 lists the unit cell
parameters for the structurally characterized compounds 1–14.
All final CIF files were checked using the CheckCIF program
(http://www.iucr.org/). Additional information concerning the
data collection and final structural solutions can be found by
accessing CIF files through the Cambridge Crystallographic Data
Base. Additional information concerning the data collection and fi-
nal structural solutions can be found in the CCDC database.
[Sm(DBP)2(
l
-O2C-DBP)]2ꢂtol (9): Used 2 (0.61 g, 0.79 mmol) in
ꢀ15 mL toluene. Solution changed from bright yellow to a green
upon CO2(g) addition. Yield 0.55 g (81%). FT-IR (KBr, cmꢁ1):
3642(m), 2962(s), 1626(s), 1589(w), 1482(w), 1413(m), 1387(m),
1370(s, sh), 1347(s), 1238(m), 1188(m), 1144(s), 1115(s),
1023(m), 867(s), 844(s, sh), 820(s), 806(s, sh), 797(s), 749(s),
715(m), 569(s), 547(m), 450(m). Anal. Calc.
C86H126O10Sm2
(MW = 1620.74 g/mol w/o toluene): C, 63.73; H, 7.84. Found: C,
63.97; H, 7.86%.
[Dy(DBP)2(
l
-O2C-DBP)]2ꢂtol (10): Used 3 (0.62 g, 0.80 mmol) in
ꢀ15 mL toluene with no color change noted. Yield 0.45 g (65%).
FT-IR (KBr, cmꢁ1): 3642(m), 2961(m), 2284(w), 1619(m), 1583(s,
sh), 1481(w), 1425(s). 1370(m, sh), 1348(m), 1263(m), 1231(s,
sh), 1188(m), 1144(m, sh), 1117(s), 1023(w), 873(s), 844(s),
822(s), 796(s, sh), 770(s), 748(s), 694(w), 663(m), 451(w). Anal.
Calc. C93H133O10Dy2 (MW = 1734.99 g/mol): C, 64.32; H, 7.73.
Found: C, 64.51; H, 8.01%.
2.4. Specific problems related to structures
For the structural solutions of 1–6, the space group changes
from P21 for Er (and all those proceeding) to P21/c (for Tm and
all those following) without a change in the overall structure. Dis-
order in the Ln centers were noted in the solutions of 1–6 [95/5%
for Ce2/Ce2a; 65/45% for Y1/Y1a; 55/45% for Er1/Er2; 50/50% for-
Yb1/Yb2; and 52/48% for Lu1/Lu2] and were modeled appropri-
ately. In addition, each structure presented a rotational disorder
in one of the t-butyl groups of the DBP that was modeled appropri-
ately. For 9 a disordered molecule of toluene was removed by the
SQUEEZE method using the PLATON program [13]. Compound 10 was
found to be ‘twinned’ and PLATON was used to create the HKLF5 for-
mat file that was then used for final refinement. Unresolved disor-
der was present in the ligands of compound 11, which did not
allow for full convergence to be reached. A higher quality data
set for these crystals could not be achieved and therefore only
the unit cell information is listed; however, it is of note that this
structure is isostructural with 13 (based on the unit cell determi-
nation and the other characterization methods). Compound 12
was found to be a true non-merohedral twin and was refined
accordingly.
[Y(DBP)2(l-O2C-DBP)]2 (11): Used 4 (0.62 g, 0.80 mmol) in
ꢀ15 mL toluene with no color change noted. Yield 0.53 g (71%).
FT-IR (KBr, cmꢁ1) 3642(s), 3080(s), 2957(s), 2872(w,sh), 1625(s),
1561(m), 1468(w,sh), 1425(s), 1389(s), 1365(s,sh), 1315(s),
1250(s), 1230(s), 1195(s), 1169(s), 1147(s), 1122(s), 1094(s),
1023(m), 879(s), 844(s), 806(s,sh), 795(s), 749(s), 620(s), 588(s),
452(s). Anal. Calc. C80H126O10Y2 (MW = 1425.66 g/mol): C, 67.38;
H, 8.91. Found: C, 67.07; H, 8.82%.
[Er(DBP)2(l-O2C-DBP)]2 (12): Used 5 (0.62 g, 0.80 mmol) in
ꢀ15 mL toluene with no color change noted. Yield 0.35 g (46%).
FT-IR (KBr, cmꢁ1): 3650(s), 3082(m,sh), 2959(s), 2874(w,sh),
2346(w), 1630(s), 1535(m), 1466(m), 1459(s), 1426(s,sh),
1413(m), 1369(s), 1347(s), 1261(s), 1231(s), 1194(s), 1143(s),
1116(s), 1094(m,sh), 1023(m), 976(m), 872(s), 844(m), 822(s,sh),
806(m), 796(m), 747(s), 710(m), 677(s), 660(m,sh), 588(m),
448(m). Anal. Calc.
60.70; H, 8.03. Found: C, 60.82; H, 7.69%.
[Yb(DBP)2( -O2C-DBP)]2 (13): Used 6 (0.63 g, 0.80 mmol) in
C80H126O10Er2 (MW = 1582.36 g/mol): C,
l
ꢀ15 mL toluene with no color change noted. Yield 0.42 g (63%).
FT-IR (KBr, cmꢁ1): 3650(m), 2961(m), 2874(w, sh), 2347(w),
1630(s), 1585(s, sh), 1420(s), 1388(m), 1372(s), 1353(m),
1262(m), 1246(m, sh), 1187(s, sh), 1143(m), 1118(s), 1024(m),
875(s), 844(s), 822(s), 804(s), 771(s, sh), 752(s), 718(s), 665(s),
590(m), 452(s). Anal. Calc. C86H126O10Yb2 (MW = 1665.01 g/mol):
C, 61.98; H, 7.63. Found: C, 62.24; H, 7.87.
3. Results and discussion
Upon insertion into a metal-ligand bond, the CO2 moiety can act
in both nucleophilic and electrophilic manners. This behavior al-
lows for a wide range of binding environments, several of which
have been crystallographically characterized for transition metals
including chelating (c), bridging (l), and chelating bridging (lc)
[Lu(DBP)2 (
l
-O2C-DBP)]2 (14): Used 7 (0.63 g, 0.80 mmol) in
modes. However, essentially no information is available for the
CO2(g) modification of [Ln(OR)3]. One report by Bochkarev et al.
[7] presented a study concerning the insertion of CO2(g) into the
supposed monomeric ‘‘Ln(OBut)3’’ (Ln = Pr, Nd, Sm) complex [7].
Infrared spectroscopy confirmed the presence of a C@O stretch
(1580 cmꢁ1) in the product, which was reported as [Ln(O2COBut)3].
Since this original study, more recent crystal structures of
‘‘Ln(OBut)3’’ compounds have been shown to adopt a number of
ꢀ15 mL toluene with no color change noted. Yield 0.36 g (54%).
FT-IR (KBr, cmꢁ1): 3648(m), 3082(w, sh), 2959(m), 2202(w),
1458(s), 1369(s, sh), 1349(m), 1316(s, sh), 1251(m, sh), 1230(s),
1194(s, sh), 1143(s), 1117(s), 1094(s, sh), 1023(m), 879(s),
844(m), 806(s, sh), 795(s), 746(s), 713(m), 678(m), 588(m),
446(m), 419(m). Anal. Calc. for C86H126Lu2O10 (MW = 1669.88 g/
mol): C, 61.86; H, 7.61. Found: C, 62.32; H, 8.02%.
nuclearities including (i) tri-nuclear {[Ln3(
(OBut)4(HOBut)2] (Ln = Ce or Dy) or [La3(
(OBut)6] containing coordinated tert-butanol}, (ii) tetra-nuclear
[Ce4(
3-OBut)3( -OBut)4(OBut)5], and (iii) penta-nuclear [Ln5(l5
O)(
3-OBut)4( -OBut)4(OBut)5] (Ln = La, Nd) [3,14]. Therefore, the
l
l
3-OBut)2(
l
2-OBut)3
l
-OBut)3
2.3. General X-ray crystal structure information
3-HOBut)2(
Single crystals were mounted onto a glass fiber from a pool of
Fluorolube™ and immediately placed in a cold N2 vapor stream,
on a Bruker AXS diffractometer employing an incident-beam
l
l
-
l
l
conclusion from the Bochkarev study was compromised due to
graphite monochromator, Mo K
a radiation (k = 0.7107 Å) and a
the erroneously-characterized starting materials.
SMART APEX CCD detector. Lattice determination and data collection
were carried out using SMART Version 5.054 software. Data reduc-
tion was performed using SAINTPLUS Version 6.01 software and cor-
rected for absorption using the SADABS program within the SAINT
software package [12]. Structures were solved by direct methods
that yielded the heavy atoms, along with a number of the lighter
atoms or by using the PATTERSON method, which yielded the
Since the reaction products of CO2(g) inserted [Ln(OR)3] com-
pounds have not been positively identified [15], we undertook
the preparation and characterization of these compounds, with
an emphasis necessarily placed on single crystal X-ray diffraction.
The H-DBP ligand was selected since it yields monomeric
[Ln(DBP)3] complexes [3,9,10], which was of interest to ensure
maximum reactivity with CO2(g) [3,9]. The synthetic route (Eq.

Leigh Anna M. Steele et al. / Polyhedron 42 (2012) 258–264
261
Table 1
Data collection parameters for 1–14.
Compound
1
2
3 [3]
4
5
6
7
Formula
Formula weight
T (K)
C
42H63CeO3
C42H63O3Sm
766.36
173(2)
C84H126Dy2O6
1556.85
173(2)
C42H63O3Y
704.83
173(2)
C42H63ErO3
783.18
173(2)
C42H63O3Yb
788.96
173(2)
C42H63Lu O3
790.89
173(2)
756.04
173(2)
Space group
monoclinic,
P2(1)
monoclinic,
P2(1)
monoclinic,
P2(1)
monoclinic,
P2(1)/c
monoclinic,
P2(1)/c
monoclinic,
P2(1)/c
monoclinic,
P2(1)/c
a (Å)
b (Å)
c (Å)
11.3190(6)
31.7246(16)
11.6835(6)
105.0760(10)
4051.0(4)
4
11.2819(7)
31.781(2)
11.6196(7)
105.0140(10)
4024.0(4)
4
11.2323(8)
31.769(2)
11.5480(8)
104.9970(10)
3980.4(5)
2
11.1912(12)
31.914(4)
11.6345(13)
104.5890(10)
4021.3(8)
4
11.1850(7)
31.902(2)
11.6241(8)
104.6790(10)
4012.3(5)
4
11.1268(7)
31.959(2)
11.6466(7)
104.3480(10)
4012.4(4)
4
11.1123(6)
31.9676(17)
11.5971(6)
104.3590(10)
3991.0(4)
4
b (°)
V (Å3)
Z
Dcalc (Mg/m3)
1.240
1.265
1.299
1.164
1.297
1.306
1.316
l
(Mo, K
a
)
)
1.157
1.493
1.911
1.484
2.125
2.365
2.508
(mmꢁ1
b
R1 (%) (all data)
3.10(3.38)
3.16(3.44)
7.76(8.31)
4.66(5.62)
8.40(8.87)
4.61(8.12)
10.43(11.20)
3.89 (6.77)
10.70(13.40)
4.42(9.10)
7.62(8.99)
3.94(4.76)
10.82(11.61)
wR2c (%) (all data) 7.88(8.32)
Compound
8
9ꢂtol
10ꢂtol
11a
12ꢂtol
13
14
Formula
Formula weight
T (K)
C
86H126Ce2O10
C86H126O10Sm2ꢂC7H8 C86H126Dy2O10ꢂC7H8 C86H126O10Y2
C86H126Er2O10ꢂC7H8 C86H126O10Yb2
C86H126Lu2O10
1669.81
173(2)
1600.11
173(2)
1712.70
1829.14
173(2)
1425.66
1838.66
1665.95
173(2)
173(2)
173(2)
173(2)
Space group
triclinic,
P1
monoclinic,
P2(1)/c
monoclinic,
P2(1)/c
monoclinic,
P2(1)/n
monoclinic,
P2(1)/c
monoclinic,
P2(1)/n
monoclinic,
P2(1)/n
ꢀ
a (Å)
b (Å)
c (Å)
12.470(3)
13.150(3)
13.646(3)
69.907(6)
75.781(5)
76.371(5)
2008.9(9)
1
25.7656(18)
17.6363(12)
21.1737(14)
25.62(2)
17.604(17)
20.99(2)
14.643(8)
14.845(9)
20.195(12)
25.6176(18)
17.6817(12)
20.9385(15)
14.5936(17)
14.8220(17)
20.045(2)
14.6015(12)
14.8211(12)
20.0160(16)
a
(°)
b (°)
95.9760(10)
95.867(12)
108.653(12)
4159(4)
95.6263(11)
108.227(2)
108.0680(10)
c
(°)
V (Å3)
9569.2(11)
4
1.189
1.266
9419(16)
4
1.290
1.629
9438.7(11)
4
1.294
1821
4118.3(8)
2
1.343
2.311
4118.1(6)
2
1.347
2.438
Z
Dcalc (Mg/m3)
1.323
1.174
l
(Mo, K
a
)
)
(mmꢁ1
b
R1 (%) (all data)
8.82(18.45)
4.93(7.17)
15.28(16.48)
7.63(10.33)
17.88 (19.50)
6.13(8.05)
12.15 (13.05)
6.94(12.62)
14.76(17.27)
5.07(9.51)
11.85 (14.77
wR2c (%) (all data) 11.72 (14.85)
a
b
c
Only the unit cell dimensions are reported due to low quality data set compound.
P
P
R1
=
||Fo| ꢁ |Fc||/ |Fo| ꢄ 100.
P
P
2
2 2 1=2
wR2 ¼ ½ wðF2o ꢁ Fc2Þ = ðwjFoj Þ ꢃ ꢄ 100.
(1)) for Ln(OR)3 was conducted in a non-polar solvent and crystals
were grown by slow evaporation. The FT-IR spectra of the powders
of 1–7 isolated from Eq. (1) revealed the absence of stretches
attributable to the –NR2 and –OH functionalities, indicating that
the amide-alcohol exchange reaction had gone to completion
[16]. To further characterize the products, the crystalline products
were analyzed by single crystal X-ray diffraction.
½LnðNðSiMe3Þ2Þ3ꢃ þ 3H-DBP ! ½LnðDBPÞ3ꢃ þ 3H-NðSiMe3Þ2
ð1Þ
As mentioned above, compounds 1 and 3 have been previously
found to adopt a trigonal geometry around the metal center [3,9].
The refinement value for our [Ce(DBP)3] structure is slightly lower
than the previously published data and so it has been included in
this report for completeness. Compounds 2, 4, 6 (Fig. 1), and 7 were
propitiously found to adopt the same structural arrangement re-
ported for the literature species [3,9,10]. Selected bond distances
and angles are listed in Table 2 and were found to be consistent
with each other and with the literature reports [3,9,10]. Upon
crystallization of
5
a
disordered oxo species [Er3(l3-O)
(l-O)3(HDBP)2(DBP)4] with cell parameters of a: 27.09, b: 13.58,
c: 28.09, b: 91.00, V = 1033 Å3 was also isolated from a slow evap-
oration over an extended period of time; however, rapid crystalli-
zation from a more concentrated reaction system led to the desired
[Er(DBP)3] (5) complex. For 1–7, the location of the Ln center was
found to be disordered, which was manifested as displacement of
the metal center from the plane. The observed displacements of
the Ln cations are reported in the metrical data Table 2. The ele-
mental analysis values for all of these compounds were found to
be within the values suggested by the crystal structure data.
Fig. 1. The [Ln(DBP)3] representative structure plot. Shown is [Yb(DBP)3].
The 13C{1H} NMR spectra were obtained for compounds 1–7 to
analyze their solution behavior. Table 3 lists the resonances for the
reference ligand H-DBP and the observed resonances for com-
pounds 1–7 (note: the paramagnetic nature of compound 5 was
too large to allow for a meaningful spectrum to be collected). For

262
Leigh Anna M. Steele et al. / Polyhedron 42 (2012) 258–264
Table 2
Metrical data for 1–14.
and ultimately proved to be a series of mono-inserted alkoxy lan-
thanide carbonates (8–14). For compound 11, the low quality of
the crystal data prevented structural convergence; however, the
basic connectivity of 11 was unequivocally established and found
to be in agreement with the other structures reported. The elemen-
tal analyses of the bulk powders of 8–14 were also found to be in
agreement with the calculated values for the crystal structure anal-
yses. For compounds 9, 10, and 12, one loosely bound toluene sol-
vent molecule had to be removed to obtain an acceptable analysis.
Compound
Ln–OR (Å)
Ln-Lna [9]
OR–Ln–OR (°)
Ce (1)
Sm (2)
Dy (3)
Y (4)
Er (5)
Yb (6)
Lu (7)
2.16
2.11
2.06
2.05
2.00
2.02
2.02
1.458
1.317
—
1.080
1.205
0.949
1.151
110.19
109.82
110.58
113.18
110.80
113.76
111.18
½LnðDBPÞ3ꢃ þ CO2ðgÞ ! ½LnðO2C-DBPÞðDBPÞ2ꢃ2
ð2Þ
Compound Ln(OR) (Å) Ln–OCOR (Å) OR–Ln–OR (°) O–Ln–O (°)
O–C–O (°)
Ce (8)
2.13
2.52
233.2
Bite 49.90
120.40
Bridge 111.80
96.40
92.98
92.46
95.09
Starting DBP complex
Ln
Inserted product
Sm (9)
Dy (10)
Y (11)
Er (12)
Yb (13)
Lu (14)
2.10
2.05
2.06
2.02
2.02
2.01
2.30
2.26
2.19
2.19
2.16
2.15
108.09
113.10
117.75
108.12
119.10
118.33
120.00
119.99
119.97
119.99
119.95
120.00
1
2
3
4
5
6
7
Ce
Sm
Dy
Y
Er
Yb
Lu
8
9
91.80
92.63
10
11
12
13
14
a
The displacement between the two disordered central Ln atoms.
these compounds more resonances than would be expected for the
threefold D3 symmetric [Ln(DBP)3] solid state structure were ob-
served in the t-butyl region. This implies either the compounds
were not pure or the methyls of the DBP t-butyl groups were
inequivalent. The former seems unlikely in that the elemental
analyses of these compounds were in agreement with the expected
values. Therefore, a ‘locked-out’ t-butyl group if persistent in solu-
tion should also be observed in the solid state. Re-examination of
the solid-state structures of 1–7 did reveal that one of the methyls
of the t-butyl groups of the DBP had a closer interaction with the Ln
metal center than the others. This interaction resulted in a distance
of r(Ln-C) av. 3.1 Å. It appears that the ‘locked-out’ t-butyl groups
(possibly occurring through H-bonding) are maintained in solu-
tion, thereby forming unique NMR environments for each methyl
carbon. This was further investigated and verified by variable tem-
perature 13C{1H} and DEPT135 NMR experiments (see Supporting
Information). Combined, the NMR data confirmed the monomeric
solid state structures were retained in solution with the presence
of a DBP methyl–lanthanide interaction.
Two structure types were observed for the CO2(g) inserted products.
The first type was noted for compound 8, which was solved as a di-
mer with each Ce atom being 5-coordinate and adopting a distorted
trigonal bipyramidal (s = 0.80) [17] geometry due to the lc-O2C-
DBP ligand (Fig. 2). The arrangement is dictated by the chelating/
bridging ligand which causes a degree of rigidity in the final confor-
mation. The average bond distances and angles of interest for 8 are
tabulated in Table 2. This
lc mode has been seen previously for the
Ce-carbamate compound Ce4(
l
4-DIPC)( 3-DIPC)( 2-DIPC)5( c-
l
l
l
DIPC)2(DIPC)3 where DIPC = N,N0-di-isopropylcarbamato) with bite,
bridge, and O–C–N angles of 50.8°, 101.5°, and 120.6° respectively
[18]. For compound 8 these angles were found to be similar, adopt-
ing 49.6° (bite), 111.8° (bridge), and 120.4° (O–C–O angle). There
are no lanthanide carbonates that contain a bridging carbonate
group; however, several transition metal species with similar car-
bonate ligands are available for comparison [average literature va-
lue of 119.9° (O–C–O)] [6b].
The remainder of the compounds (9–14) formed the same gen-
eral structure motif shown in Fig. 3 (compound 13 is used as a rep-
resentative compound). In this arrangement each Ln cation adopts
a distorted tetrahedral (Td) geometry using two terminal DBP and
3.1. CO2(g) insertion reactions
With this set of monomeric starting compounds (1–7) available,
CO2(g) insertion reactions were undertaken as shown in Eq. (2).
The resultant products were isolated as X-ray quality crystals
two bridging
l-O2C-DBP ligands (Fig. 3). For 9, 10, and 12, two
independent molecules were found in the unit cell, and one tolu-
ene molecule was also located in the lattices of 9, 10 and 12. The
carbonate bond angles listed in Table 2 for 9–14 are in agreement
with what has been reported for carbonates found in transition
metal complexes (O–C–O av. 119.9°). The change in bonding mode
from 8 is associated with the size of the Ln cation.
Quite surprisingly, the products crystallized were consistent
with the insertion of only one CO2 per Ln atom. The mother liquor
of the reaction mixture was not analyzed for other potential prod-
ucts since crystallographically identified species were of interest
for this study; however, there is no inherent reason to expect only
one CO2(g) to insert into these Ln(DBP)3 complexes. To explore
whether the compound could insert additional CO2 molecules into
the other remaining Ln–O bonds, elevated pressures were used to
attempt to force multiple CO2 insertions into 3. Reactions were run
under similar experimental conditions expect the CO2(g) pressures
were raised as high as 90 psig. Using the Dy analogue as an exam-
ple, after stirring 3 with high pressure CO2(g) for 30 min the color-
less solution had turned dark brown and was kept at these
conditions for a total of 1 h, with no further color change noticed.
Table 3
13C{1H} NMR chemical shifts for 1–4 and 6–7.
Compound
d
13C{1H}: phenyl carbonsa
13C{1H}: t-butyl carbons
–C(CH3)3
155.0 135.9 125.0 120.3 34.3
–C(CH3)3
H-DBP
30.3
32.3
30.3
31.3
30.3
b
1
2
[Ce(DBP)3]
155.9 135.9
—
120.2 37.1
34.3
120.3 36.2
118.7 34.3
30.2
[Sm(DBP)3] 154.2 138.8
—
135.9
3
4
[Dy(DBP)3]
[Y(DBP)3]
154.2 135.9
154.2 137.5
136.0
—
120.2 34.3
120.3 35.4
117.6 34.3
120.3 34.3
120.3 35.0
118.4
30.4
32.6
30.3
30.4
32.2
30.4
6
7
[Yb(DBP)3]
[Lu(DBP)3]
154.3 135.8
154.2 135.8
—
—
a
13C{1H} NMR shifts for 1–7. Shifts in bold have been confirmed to be the correct
assignments for each carbon by DEPT135 experiments.
b
— resonances could not be separated from the tol-d8 resonances.

Leigh Anna M. Steele et al. / Polyhedron 42 (2012) 258–264
263
because the packing disorders that alkoxides exhibit based on re-
stricted rotation would yield erroneous peaks and limit the amount
of useful information for the compounds [14a]. FT-IR analyses indi-
cated that the C@O stretch of –O2COR was present as shown by a
sharp band in the range of 1640–1620 cmꢁ1: 8 (1630 cmꢁ1), 9
(1626); 10 (1619), 11 (1625), 12 (1630), 13 (1630), 14 (1636). If a
l
- or lc-O2C bonding mode was present, a second weaker CO vibra-
tional band in the range of 1355–1335 cmꢁ1 should be present in
the IR spectra as has been observed for similar compounds with
transition metals, and these bands are seen [16,20]:
8
(1346 cmꢁ1), 9 (1347), 10 (1348), 11 (1361), 12 (1347), 13 (1353),
and 14 (1349). This confirms the bonding modes of the carbonato
groups for 8–14 are present in the bulk powder.
4. Conclusions
The insertion of CO2(g) into Ln–O bonds of the monomeric fam-
ily of [Ln(DBP)3] (1–7) compounds was found, for the first time, to
form either the mono-inserted [Ce(lc-O2C-DBP)(DBP)2]2 (8) or
[Ln( -O2C-DBP)(DBP)2]2 (9–14) species. The only substantive vari-
l
ation in these compounds is the bonding mode of the carbonato
moiety, binding in a bridging chelate mode for the larger Ce atom
in 8 and in simple bridging modes for the smaller lanthanides (9–
14). Surprisingly, for the under-coordinated Ln metal centers only
one of the three Ln–OR bonds was found to be reactive, even at
higher CO2(g) pressures. This is preliminarily attributed to the ste-
ric hindrance and t-butyl interaction of the DBP ligands with the Ln
center reducing the number of sites for CO2(g) to insert. Further
exploration of this effort currently is underway exploring less ste-
rically demanding ligands as well as electronically varied ligands
(e.g., siloxides).
Fig. 2. Structure plot of 8. Thermal ellipsoids of the heavy atoms are drawn at 30%
level. Carbons are drawn as ball and stick for clarity.
Acknowledgments
This work was supported by the Laboratory Directed Research
and Development (LDRD, program 151300) program at Sandia
National Laboratories and the National Science Foundation (Grant
CHE09-11110 to R.A.K.). The Bruker X-ray diffractometer was
purchased via a National Science Foundation CRIF:MU award to
the University of New Mexico (CHE04-43580). Sandia National
Laboratories is a multi-program laboratory managed and operated
by Sandia Corporation, a wholly owned subsidiary of Lockheed
Martin Corporation, for the U.S. Department of Energy’s National
Nuclear Security Administration under contract DE-AC04-
94AL85000.
Fig. 3. Structure plot of 13. Thermal ellipsoids of the heavy atoms are drawn at 30%
level. Carbons are drawn as ball and stick for clarity.
Crystals were obtained upon slow evaporation of the solution and
yielded an identical structure previously seen as 10. This selectivity
is of interest since Ln cations are known to have high coordination
numbers (i.e., 7 and 8) and the final coordination geometry noted
was only pseudo-Td or tbp geometry. We have preliminarily attrib-
uted this to steric encumbrance from the t-butyl groups’ interac-
tions with the Ln metal center, as noted in the NMR and crystal
structure results (vide infra). These interactions may fill open sites
and prevent additional CO2(g) insertions. Further efforts with
smaller and varied sterically hindering ligands are underway to ex-
plore this phenomenon and possible control the sites of multiple
CO2 insertions.
Studies were undertaken to further elucidate the carbonate
binding modes. The 13C NMR resonance for carbonate ligands is re-
ported to be near d 190.0 ppm [19]. The expected slow relaxation
times of Ln species, coupled with the limited solubility of the
CO2(g) insertion products (8–14), made obtaining useful NMR spec-
tra difficult even with extended data collection times. Solid state
NMR spectroscopy was not performed for any of the compounds
Appendix A. Supplementary data
CCDC 853343–853354 and 879692 contain the supplementary
crystallographic data for 1–2, 4–7, and 8–14. These data can be ob-
tained free of charge via http://www.ccdc.cam.ac.uk/conts/retriev-
ing.html, or from the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033;
or e-mail: deposit@ccdc.cam.ac.uk. Supplementary data associated
with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.poly.2012.05.021.
References
[1] (a) T.J. Boyle, H.D. Pratt, B.A. Hernandez-Sanchez, T.N. Lambert, T.J. Headley, J.
Mater. Sci. 42 (2007) 2792;
(b) T.J. Boyle, S.D. Bunge, N.A. Andrews, L.E. Matzen, K. Sieg, M.A. Rodriguez,
T.J. Headley, Chem. Mater. 17 (2004) 3279.
[2] B.E. Douglas, J. Chem. Educ. 31 (1954).
[3] T.J. Boyle, S.D. Bunge, P.G. Clem, J. Richardson, J.T. Dawley, L.A.M. Ottley, M.A.
Rodriguez, B.A. Tuttle, G.R. Avilucea, R.G. Tissot, Inorg. Chem. 44 (2005) 1588.

264
Leigh Anna M. Steele et al. / Polyhedron 42 (2012) 258–264
[4] T.J. Boyle, L.A.M. Ottley, T.M. Alam, M.A. Rodriguez, P. Yang, S.K. McIntyre,
[13] A.L. Spek, J. Appl. Crystallogr. 46 (2006) 7.
Polyhedron 28 (2010) 1784.
[14] (a) T.J. Boyle, L.J. Tribby, S.D. Bunge, Eur. J. Inorg. Chem. (2006) 4553;
(b) D.C. Bradley, H. Chudzynska, M.B. Hursthouse, M. Motevalli, Polyhedron 10
(1991) 1049;
[5] T. Mioduski, J. Radioanal. Nucl. Chem., Akadémiai Kiadó, vol. 179, Co-published
with Springer Science+Business Media B.V., Formerly Kluwer Academic
Publishers B.V., 1993, p. 371.
[6] (a) T.J. Boyle, L.A.M. Ottley, Chem. Rev. 108 (2008) 1896;
(b) CCDC ConQuest In ConQuest version 1.13 Cambridge Crystallographic Data
Center: www.ccdc.cam.ac.uk, 2011 ed., vol. 2.32, 2011.
(c) S. Daniele, L.G. Hubert-Pfalzgraf, P.B. Hitchcock, M.F. Lappert, Inorg. Chem.
Commun. 3 (2000) 218;
(d) J. Gromada, A. Mortreux, T. Chenal, J.W. Ziller, F. Leising, J.F. Carpentier,
Eur. J. Inorg. Chem. 8 (2002) 3773.
[7] M.N. Bochkarev, E.A. Fedorova, Y.F. Rad’kov, G.S. Kalinina, S.Y. Khorshev, G.A.
Razuvaev, Dokl. Akad. Nauk. SSSR 279 (1984) 1386.
[15] A. Behr, Angew. Chem., Int. Ed. 27 (1988) 661.
[16] G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and
Charts, third ed., Wiley, West Sussex, 2001.
[17] A.W. Addison, T. Nageswara Rao, J. Reedijk, J. Van Rijn, G.C. Verschoor, J. Chem.
Soc., Dalton Trans. (1984) 1349.
[18] U. Baisch, D.B. Dell’Amico, F. Calderazzo, L. Labella, F. Marchetti, D. Vitali, J.
Mol. Catal. A 204 (2003) 259.
[19] M.G. Gardiner, N.W. Davies, A.S.P. Frey, J. Wang, Chem. Commun. (2006)
4853.
[8] X. Zhou, M. Zhu, J. Organomet. Chem. 647 (2001) 28.
[9] H.A. Stecher, A. Sen, A.L. Rheingold, Inorg. Chem. 27 (1988) 1130.
[10] H. Amberger, H. Reddmann, C. Guttenberger, B. Unrecht, L. Zhang, C. Apostolidis,
O. Walter, B. Kanellakopulos, Z. Anorg. Allg. Chem. 629 (2003) 1522.
[11] (a) E.D. Bradley, D.L. Clark, J.C. Gordon, P.J. Hay, W. Keogh, R. Poli, B.L. Scott,
J.G. Watkin, J. Inorg. Chem. 42 (2003) 6682;
(b) W.A. Herrmann, R. Andwander, F.C. Munck, W. Scherer, V. Dufaud, N.W.
Huber, G.R. Artus, Z. Naturforsch. B 49 (1994) 1789;
(c) M. Niemeyer, Z. Anorg. Allg. Chem. 628 (2002) 647;
[20] (a) P. Chaudhuri, V. Pavlishchuk, F. Birkelbach, T. Weyhermuller, K. Wieghardt,
Inorg. Chem. 41 (2002) 4405;
(d) W.S. Rees Jr., O. Just, D.S. Van Derveer, J. Mater. Chem. 9 (1999) 249;
(e) G. Secarel, C. Wiemer, M. Fanciulli, I.L. Fedushkin, G.K. Fukin, G.A.
Domrachev, Y. Lebedinskii, A. Zenkevich, G. Pavia, Z. Anorg. Allg. Chem. 633
(2007) 2097;
(b) A.G. Samuelson, R. Ghosh, M. Nethaji, Chem. Commun. (2003) 2556;
(c) T.-B. Lu, J.-M. Chen, W. Wei, X.-L. Feng, J. Asian Chem. 2 (2007) 710.
(f) M. Westerhausen, M. Hartmann, A. Pfitzner, W. Schwarz, Z. Anorg. Allg.
Chem. 621 (1995) 837.
[12] (a) G.M. Sheldrick, Acta Crystallogr., Sect. A A64 (2008) 112;
(b) G.M. Sheldrick, SADABS (2.01), Bruker/Siemens Area Detector Absorption
Correction Program, Bruker AXS, Madison, Wisconsin, USA, 1998.