Tetramethylcyclopentadienyl Ligands Allow Isolation of Ln(II) Ions across the Lanthanide Series in [K(2.2.2-cryptand)][(C5Me4H)3Ln] Complexes

Article abstract of DOI:10.1021/acs.organomet.8b00557

Although previous studies of the stabilization of Ln(II) ions across the lanthanide series have relied on Me₃Si-substituted cyclopentadienyl ligands, we now find surprisingly that these ions can also exist surrounded by three tetramethylcyclope

Full text of DOI:10.1021/acs.organomet.8b00557

Article
pubs.acs.org/Organometallics
Cite This: Organometallics XXXX, XXX, XXX−XXX
Tetramethylcyclopentadienyl Ligands Allow Isolation of Ln(II) Ions
across the Lanthanide Series in [K(2.2.2-cryptand)][(C5Me4H)3Ln]
Complexes
Tener F. Jenkins, David H. Woen, Luke N. Mohanam, Joseph W. Ziller, Filipp Furche,*
and William J. Evans*
Department of Chemistry, University of California, Irvine, California 92697-2025, United States
S
* Supporting Information
ABSTRACT: Although previous studies of the stabilization
of Ln(II) ions across the lanthanide series have relied on
Me3Si-substituted cyclopentadienyl ligands, we now find
surprisingly that these ions can also exist surrounded by
three tetramethylcyclopentadienyl ligands. Reduction of the
4fn Ln(III) complexes, Cptet3Ln (Cptet = C5Me4H) using
potassium graphite in the presence of 2.2.2-cryptand (crypt)
produces the Ln(II) complexes, [K(crypt)][Cptet3Ln] for Ln
= La, Ce, Pr, Nd, Sm, Gd, Tb, and Dy, all of which were
characterized by X-ray crystallography. These complexes
display intense absorptions in the UV−visible−near IR region
that are red-shifted compared to those of previously
characterized (Cp′3Ln)1− complexes (Cp′ = C5H4SiMe3). The thermal stability of these new Ln(II) complexes decreases
with the size of the metal.
C5H2(CMe3)3 (Cpttt)20,21 and the tris(aryloxide) mesitylene
INTRODUCTION
In recent years, the range of oxidation states available to the
■
ligand, [(Ad,MeArO)3mes]3−.22,23 Extensive crystallographic,
spectroscopic, magnetic, and density functional theory
(DFT) studies showed that the new Ln(II) ions in the
tris(cyclopentadienyl) environments adopted 4fn5d1 electron
configurations.1,7−12,23−26 This could be rationalized by the
crystal field splitting of a tris(cyclopentadienyl) coordination
rare-earth metals in crystallographically characterizable molec-
ular complexes available for reactivity in solution has greatly
expanded.1,2 Up until 2001, it was thought that only six
lanthanides could form crystallographically characterizable
molecular complexes of Ln(II) ions in solution: Eu, Yb, Sm,
Tm, Dy, and Nd.3−6 These complexes could be made by
reduction of 4fn Ln(III) precursors and formed Ln(II) ions
with 4fn+1 electron configurations as expected. However, it is
now known that yttrium and all of the lanthanides (except Pm
which was not studied due to its radioactivity) can form
isolable molecular complexes of Ln(II) ions if reductions are
done in the proper coordination environment.7−12 Specifically,
reduction of tris(cyclopentadienyl) complexes with silyl-
substituted ligands C5H3(SiMe3)2 (Cp″)7,8 and C5H4SiMe3
(Cp′)9−12 provided access to Ln(II) ions across the series, as
shown in eq 1.1,2 This was also extended to the actinides, Th,13
2
environment, which puts a 5dz orbital comparable in energy to
the 4f orbitals.27−32
A comparison of the rare-earth metals in the [K(crypt)]-
[Cp′3Ln] series (crypt = 2.2.2-cryptand) revealed that Ln(II)
ions could be grouped into three categories.10 Sm, Eu, Tm, and
Yb form complexes of traditional Ln(II) ions with 4fn+1
electron configurations. For Ln = Nd and Dy, the [K(crypt)]-
[Cp′3Ln] complexes have 4fn5d1 configurations, but in other
ligand environments, the metals form 4fn+1 Ln(II) ions. These
are designated as configurational crossover ions. Nd(II) and
Pr(II) are known to be configurational crossover ions in the
solid state iodides, LnI2.33,34 The third category contains the
rest of the lanthanide metals which have 4fn5d1 configurations
and Y(II) which is a 4d1 ion.
After the first Y(II) complex, (Cp′3Y)1−, was isolated,11 the
importance of the ligand system was examined by studying the
reductions of Cp3Y(THF) (Cp = C5H5), CpMe3Y(THF) (CpMe
= C5H4Me), and Cp″3Y.25 In each case, the reduction product
U,14,15 Pu,16 and Np.17 Examples of complexes of new Ln(II)
ions are also known with C5H3(CMe3)2 (Cptt)18,19 and
Received: August 2, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
was examined in solution by EPR spectroscopy.9,11 Each EPR
spectrum showed a two-line hyperfine pattern, which is
consistent with the interaction of an unpaired electron with
the 100% naturally abundant 89Y (I = 1/2) nucleus. However,
none of these ligand systems gave isolable Y(II) complexes as
found in the reduction of Cp′3Y. Comparison of the hyperfine
coupling constants, Table 1, of these tris(cyclopentadienyl)
was too electron donating to form stable complexes. We report
here that the (Cptet 3− ligand set provides an entire new series
)
3
of Ln(II) complexes that allows evaluation of the three
categories of Ln(II) complexes as a function of metal.
RESULTS
■
Yttrium. The reduction of Cptet3Y, 1-Y, in THF using KC8
in the presence of crypt produced a dark blue solution, 2-Y,
that decomposed to an orange solution within minutes even at
−35 °C. In order to obtain EPR data on the reduction product,
the reaction was performed in an EPR tube at −78 °C.
The EPR spectrum, Figure 1, collected from the frozen
solution at 77 K shows an axial signal with a two-line hyperfine
pattern arising from the I = 1/2 of 100% naturally abundant
89Y, similar to those previously observed in yttrium reduction
reactions.1,25 The spectrum could be simulated to give g∥ =
2.001 and g⊥ = 1.981 with an average hyperfine coupling
constant of Aavg = 62.8 G. After the room temperature
measurement was completed, the sample had already
decomposed to a pale orange/yellow solution. The observed
62.8 G hyperfine coupling constant fits in well with the data in
Table 1 in that the value is high compared to other substituted
cyclopentadienyl ligands, which is suggestive of increasing
electron donation to the cyclopentadienide ring. Attempts to
isolate the reactive Y(II) complex were not successful.
Table 1. Comparison of EPR Parameters of Reduction
Products of YL3 Complexes (L = Ligand)
compound reduced
giso
Aiso (G)
Cp″3Y25
1.9908
1.991
1.9905
1.9903
1.976
36.1
36.6
42.8
46.9
110
Cp′3Y25
Cp3Y(THF)25
CpMe3Y(THF)25
35
Y(NR2)3
complexes along with that of the solution reduction product of
Y(NR2)3 (R = SiMe3), a system that also did not give an
isolable Y(II) complex,35 suggested an explanation for the
observed results. The more unstable complexes had larger
hyperfine coupling constants. These larger hyperfine coupling
constants observed in the reductions of Y(NR2)3, CpMe3Y, and
Cp3Y compared to Cp′3Y suggested that the NR2, CpMe, and
Cp ligands were more electron donating than Cp′ in this
yttrium system and made the Y(II) complex too electron-rich
to be isolated. These results were consistent with data obtained
on electrochemical analysis of (C5R5)2ZrCl2 complexes,36 as
well as studies by Lappert that showed that it was more
d i ffi c u l t t o r e d u c e [ C 5 H 3 ( C M e 3 ) 2 ] 3 L a t h a n
[C5H3(SiMe3)2]3La.19 The instability of Cp″3Y is attributed
to the steric crowding in this complex with the large ligand and
the small metal.37,38
However, the recent isolation of +2 ions in complexes of
NR2 (R = SiMe3) ligands suggested that strongly donating
ligands could also provide these new ions. Specifically, amide
ligands were used to isolate the first crystallographically
characterizable complex of a +2 ion of the smallest rare-earth
metal, scandium, as well as the lanthanides, Ln = Nd, Gd, Tb,
Dy, Ho, and Er, in the compounds [M(chelate)][Ln(NR2)3]
(M = K, Rb; chelate = 18-crown-6 and crypt; R = SiMe3).39,40
Consequently, we have examined the reduction of Cptet3Ln
complexes (Cptet = C5Me4H). These complexes were not
examined earlier25 because it was assumed that the Cptet ligand
Neodymium. Nd was chosen for further studies for two
reasons. Following the isolation and structural characterization
of the Pu(II) complex, [K(crypt)][Cp″3Pu] [crypt = 2.2.2-
cryptand, Cp″ = C5H3(SiMe3)2],16 it became of interest to
expand the isolation of the +2 oxidation state to another
transuranic metal, americium, with other cyclopentadienyl
ligand sets. Neodymium was chosen as an appropriate model
for this investigation, since the predicted nine-coordinate
Nd(II) ionic radius of 1.35 Å is similar to the 1.31 Å of the
predicted nine-coordinate Am(II) ionic radius.42 Nd was also
of interest, since Nd(II) is a configurational crossover ion and
could conceivably form a traditional 4fn+1 complex that would
have better stability than the “(Cptet3Y)1−” complex above. Nd
is also considerably larger than yttrium, and the steric match of
ligand set with metal size could be more favorable.
The reaction of Cptet3Nd (Cptet = C5Me4H), 1-Nd, with
KC8 in the presence of crypt in THF under argon at −35 °C
produced a dark green solution. The product was crystallized
from the THF solution by layer diffusion with Et2O at −35 °C,
Figure 1. X-band EPR spectra of the reduction product of Cptet3Y, 2-Y, collected at 77 K (left; mode: perpendicular; g∥ = 2.001, g⊥ = 1.981; Aavg
=
62.8 G; ν = 9.624 GHz; P = 2.026; modulation amplitude = 0.0407 mT) and 298 K (right; mode: perpendicular; giso = 1.986 ; Aiso = 64.8 G; ν =
9.817 GHz; P = 2.021; modulation amplitude = 2.853 mT). The simulated spectra (dotted lines) were generated with the exclusion of the extra
feature in the right spectrum attributed to an electride species.41
B
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Thermal ellipsoid plot of [K(crypt)][Cptet3Nd], 2-Nd, drawn at 50% probability level. Hydrogen atoms are omitted for clarity.
and it was identified by X-ray crystallography as the Nd(II)
complex, [K(crypt)][Cptet3Nd], 2-Nd, Figure 2, eq 1.
were grown over several days by transferring a freshly prepared
THF solution of the sample to the bottom of a cooled vial of
Et2O for slow layer diffusion. Crystallographic data were
collected on all of these samples which provided a new series
of Ln(II) ions in a single coordination environment for
comparison with the 3-Ln series.9,11,12 The Nd structure in
Figure 2 is representative of these complexes, and spectro-
scopic data are presented later. It should be noted that mixed
ether solvent occupancy was found in the lattice for Pr, Nd,
Sm, Gd, Tb, and Dy, but no solvents were found in the crystal
lattice of the complexes of the largest metals, La and Ce (see
the Supporting Information for details).
Ho and Er. Attempts to make the Ho and Er analogues also
gave dark solutions, but they did not persist for more than a
few hours even at −35 °C and were not pursued further. This
instability was consistent with that of the similarly sized
yttrium complex.
The Nd−ring centroid distances of 2-Nd range between
2.555 and 2.568 Å with an average of 2.563 Å. This is 0.045 Å
longer than the 2.518 Å Nd−ring centroid distance of the
Nd(III) precursor, 1-Nd.43 This small change from Ln(III) to
Ln(II) is consistent with that observed in going from 4fn to
4fn5d1 electron configurations with the new ions in the
[K(crypt)][Cp′3Ln], 3-Ln, series.10 In contrast, 4fn to 4fn+1
reductions with the traditional ions, Eu(II), Yb(II), Sm(II),
and Tm(II), give bond distance changes of 0.1−0.2 Å.1−6 The
small change in distances between the Nd(III) and Nd(II)
EPR Spectra. The EPR spectra of [K(crypt)][Cptet3La], 2-
La, and [K(crypt)][Cptet3Gd], 2-Gd, are shown in Figure 3.
The eight-line hyperfine pattern of 2-La is as expected for
La(II), since the 99.9% naturally abundant 139La has an I = 7/2
nuclear spin. The data are consistent with g = 1.98 and A =
291. The average hyperfine coupling constant for 2-La is larger
than the values found for 3-La,10 154 G, and [K(crypt)]-
[Cp″3La],7 133.5 G, as well as the reduction product of
CpMe3La,44 195 G, as expected for the more electron donating
Cptet ligand. The spectrum of 2-Gd is similar to that of the
crystallographically characterized 3-Gd,9 as well as the
2
complexes suggests that the added electron occupies the dz
orbital.
La, Ce, Pr, Sm, Gd, Tb, and Dy. Following the successful
isolation of 2-Nd, the syntheses of [K(crypt)][Cptet3Ln]
complexes were pursued across the lanthanide series. In the
case of Ln = La, Ce, Pr, Sm, Gd, Tb, and Dy, reduction can be
achieved with potassium graphite in the presence of crypt at
−35 °C. The resulting dark solutions are stable enough to
provide crystals of [K(crypt)][Cptet3Ln], 2-Ln. These crystals
Figure 3. Room temperature EPR spectra of [K(crypt)][Cptet3La] (left; mode: perpendicular; giso = 1.97; Aiso = 291 G; ν = 9.8175 GHz; P = 2.021;
modulation amplitude = 1 mT) and [K(crypt)][Cptet3Gd] (right; mode: perpendicular; giso = 1.9786; ν = 9.8182 GHz; P = 2.026; modulation
amplitude = 1.464 mT).
C
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
reduction products of Cp″3Gd, Cp″2GdCp, and
Cp″2GdCpMe.25
0.147 and 0.099 Å for 2-Sm and 2-Dy, respectively. This is
substantially larger than the 0.045−0.058 Å differences for the
six other metals in the series. These larger distances suggest
that 2-Sm and 2-Dy contain 4f6 and 4f10 ions, respectively, and
not 4fn5d1 species. These metrical parameters and 4fn+1
electronic configuration assignments are consistent with the
UV−vis spectra presented below. This is also reasonable, since
Sm is known as a traditional Ln(II) ion and Dy is known as a
configurational crossover ion.10 The difference in these two
structures is shown graphically in Figure 4, which plots Ln−
Structural Analysis. Metrical parameters on the [K-
(crypt)][Cptet3Ln], 2-Ln, series are presented in Table 2,
Table 2. Comparison of Ln−(Cyclopentadienyl Ring
Centroid) Distances (Ln−Cnt) for [K(crypt)][CpX Ln]
3
with CpX = Cptet, Cp′, and Cp″
Ln−Cnt range
Ln−CntAve
a
(Å)
(Å)
Δ[Ln(II) vs Ln(III)]
(Cptet3La)1−
2.626−2.642
2.606−2.642
2.581−2.595
2.594−2.612
2.574−2.609
2.553−2.567
2.572−2.583
2.552−2.588
2.530−2.544
2.555−2.568
2.530−2.559
2.514−2.528
2.623−2.640
2.603−2.615
2.511−2.519
2.463−2.475
2.498−2.505
2.448−2.461
2.543−2.543
2.434−2.450
2.633
2.620
2.586
2.603
2.587
2.558
2.578
2.566
2.535
2.563
2.544
2.519
2.630
2.608
2.516
2.468
2.502
2.454
2.543
2.443
0.05845
0.0187,46
0.02610,38
0.05147
0.0228,48
0.02910,48
0.04647
b
0.0269,49
0.04543
0.0198,46
0.03110,49
0.14745
0.14810
0.047
(Cp″3La)1−
(Cp′ La)1−
3
(Cptet3Ce)1−
(Cp″3Ce)1−
(Cp′3Ce)1−
(Cptet3Pr)1−
(Cp″3Pr)1−
(Cp′3Pr)1−
(Cptet3Nd)1−
(Cp″3Nd)1−
(Cp′3Nd)1−
(Cptet3Sm)1−
(Cp′3Sm)1−
(Cptet3Gd)1−
(Cp′3Gd)1−
(Cptet3Tb)1−
(Cp′3Tb)1−
(Cptet3Dy)1−
(Cp′3Dy)1−
Figure 4. Plot of average Ln−(Cptet centroid) distances in Cptet3Ln,
1-Ln (blue diamonds), and in 2-Ln (orange squares).
0.0319
0.05445
0.0329
0.099
0.03610
Cnt distances as a function of atomic number. In addition, the
Evans method51−53 magnetic moment of 2-Dy is 10.8 μB. This
is close to the 10.6 μB theoretical value for a 4f10 complex and
is smaller than the 11.3−11.7 μB values observed for 4f95d1
complexes.24
a
Δ[Ln(II) vs Ln(III)] = the difference in Ln−Cnt distances of
[Cpx3LnII]1− vs Cpx3LnIII. The references are to the structures of the
b
Cpx3LnIII complexes. The structure of Cp″3Pr has not been reported
UV−vis Spectroscopy. The UV−vis spectra of 2-Ln are
shown in Figure 5, and the absorbance maxima and extinction
coefficients are compared with those of 3-Ln in Table 3. The
spectra of 2-Ln for Ln = La, Ce, Pr, Nd, Gd, and Tb show the
strongest absorptions in the near-infrared region with λmax
values in the range 745−874 nm with a molar extinction
coefficient, ε, of 1700−5600 M−1 cm−1, Figure 5. In
comparison, the previously reported spectra of 3-Ln complexes
of these metals have the largest absorptions from 420 to 635
nm in the visible region with ε = 4400−6500 M−1 cm−1.1,2,9−12
In contrast to the UV−vis spectra of La, Ce, Pr, Nd, Gd, and
Tb, the spectra of 2-Dy and 2-Sm have weaker absorptions.
These are shown with a different scale in Figure 5 (right). Less
intense absorptions were previously observed for the 3-Ln
complexes for the metals with 4fn+1 electron configurations, i.e.,
Eu, Yb, Sm, and Tm.10 Hence, the λmax = 466 nm with ε = 203
M−1 cm−1 for 2-Sm is consistent with a 4f6 electron
configuration for this complex. The weaker absorption for 2-
Dy with λmax = 766 nm and ε = 200 M−1 cm−1 is also
consistent with a 4fn+1 configuration, and this matches the
structural data above which suggested that 2-Dy had a 4f10
configuration.
for comparison.8
along with data on the analogous [K(crypt)][Cp′3Ln], 3-
Ln,9−12 and [K(crypt)][Cp″3Ln]7,8 complexes when available.
The Ln−(C5Me4H ring centroid) distances (Ln−Cnt) for six
of the eight 2-Ln complexes with Ln = La, Ce, Pr, Nd, Gd, and
Tb are consistent with Ln ionic radii, decreasing with
increasing atomic number, following the lanthanide contrac-
tion. For each of these six metals, the Ln−Cnt distance also
decreases in the order: Cptet ≫ Cp″ > Cp′. This suggests that
the (Cptet
)
environment occupies more space than
3−
3
(Cp″3)3−, which is surprising given that (Cp″)1− was
investigated as a ligand for being sterically similar to
(C5Me5)1−.50
The differences in Ln−Cnt distances between the Ln(III)
(C5R5)3Ln precursor and the reduced Ln(II) product for La,
Ce, Pr, Nd, Gd, and Tb are 0.045−0.058 Å for the Cptet series
versus 0.018−0.022 Å for the Cp″ complexes and 0.026−0.031
Å for the Cp′ compounds. All of these differences are much
smaller than the 0.1−0.2 Å differences found for 4fn+1 Ln(II)
versus 4fn Ln(III) complexes of the traditional Ln(II) ions of
Eu, Yb, Sm, and Tm. This is structural evidence consistent with
4fn5d1 configurations for these six metals based on previous
structural, spectroscopic, and DFT analysis of the 3-Ln
complexes.9−12
Theoretical Analysis. The structures of the Cptet3Ln
precursors and the (Cptet3Ln)1− anions in the 2-Ln complexes
were optimized with density functional theory (DFT) using
the Tao−Perdew−Staroverov−Scuseria meta-generalized gra-
dient approximation hybrid (TPSSh) density functional along
with split valence basis sets54 for the ligands and quasi-
relativistic f-in-core effective core potentials (ECPs)55 for 2-La,
3-La, 2-Ce, 3-Ce, 2-Sm, 3-Sm, 2-Gd, and 3-Gd (see the
Supporting Information for further details); f-in-core electron
Sm and Dy. The metrical parameters of the samarium and
dysprosium complexes, 2-Sm and 2-Dy, differ from those of
the six metals above. The difference in Ln−Cnt distances
between the Ln(II) complexes and their Ln(III) precursors are
D
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 5. UV−vis spectra of 2-Ln (Ln = La, Ce, Pr, Nd, Sm, Gd, Tb, Dy) collected at 5.0 mM concentration in THF at room temperature on the
left and expanded scale spectra of 2-Dy and 2-Sm on the right.
Table 3. Comparison of UV−vis Absorption Maxima of
[K(crypt)][Cptet3Ln], 2-Ln, and [K(crypt)][Cp′3Ln], 3-Ln
λmax (nm)
ε (M−1 cm−1
)
2-La
745
554
874
635
854
518
833
420
466
360
745
430
784
464
766
483
1700
6500
4100
4700
4500
4500
5600
4700
200
3-La10
2-Ce
3-Ce10
2-Pr
3-Pr9
2-Nd
3-Nd10
2-Sm
3-Sm10
2-Gd
3-Gd9
2-Tb
700
2500
4400
650
4800
200
Figure 6. Highest singly occupied molecular orbital of 2-La (contour
value 0.04 au), obtained using DFT with the TPSSh functional and
the f-in-core SCecp-mwb-d ECP (see the Supporting Information for
details).
3-Tb9
2-Dy
3-Dy10
3400
The f-out-of-core calculations converged to a predominantly
4f4 state but with a Nd−centroid distance of 2.621 Å.
Calculations with a [4f3]5d1 configuration gave a value of 2.580
Å, closer to the metrical parameters from the X-ray crystal
structure, and excitation spectra in qualitative agreement with
experiment (see below). These results provide support for the
assignment of 2-Nd as a 4f35d1 ion with a low-lying excited 4f4
configuration.
In contrast to the results above, calculations on the
traditional Sm(II) ion with a [4f6] electron configuration in
(Cptet3Sm)1− led to a 2.696 Å, Sm−Cnt distance which is
longer than the 2.630 Å observed distance. However, f-out-of-
core calculations predicted a 4f6 configuration with a 2.610 Å
Sm−centroid distance in close agreement with experiment. For
the configurational crossover ion Dy(II), the calculated bond
distance for neither the [4f10] configuration, 2.650 Å, nor the
[4f9]5d1 configuration, 2.505 Å, matched the 2.543 Å observed
value. However, the calculated differences in Ln−ring centroid
distances of the Ln(III) Cptet3Ln and the Ln(II) (Cptet3Ln)1−
for Sm and Dy, 0.200 and 0.199 Å, respectively, were much
larger than those of the 4fn5d1 ions described above. The
experimentally observed Ln(II) vs Ln(III) differences for Sm
and Dy, 0.147 and 0.099 Å, are not as large as the calculated
differences, but the data support the presence of 4fn+1
configurations for the ions in (Cptet3Sm)1− and (Cptet3Dy)1−,
which is consistent with the UV−vis data. Time-dependent
density functional theory (TDDFT) simulations are consistent
configurations are denoted by brackets in the following
discussion. This methodology was previously established for
the 3-Ln series9 and further validated by f-out-of-core
calculations56 for 2-La, 3-La, 2-Ce, 3-Ce, 3-Pr, 2-Sm, 3-Sm,
2-Gd, and 3-Gd (see the Supporting Information for further
details).
For the La, Ce, Pr, Nd, Gd, and Tb anions in 2-Ln, the
calculated average Ln−(Cp ring centroid) distances are within
0.001−0.02 Å of the experimental values. Moreover, for these
six metals, the computed change in Ln−Cnt distances between
the Ln(III) precursor Cptet3Ln and the reduced Ln(II)
product, 2-Ln, matches the X-ray data within 0.02 Å when
[4fn]5d1 configurations for the anions are assumed with the f-
in-core effective core potentials (Table S24, Supporting
Information). The highest occupied molecular orbital of 2-
2
La is shown in Figure 6. The 5dz nature of this HOMO
matches those found for 3-Ln.
Since Nd(II) is a configurational crossover ion that has been
observed to access both 4f4 and 4f35d1 electron configurations,
calculations on (Cptet3Nd)1− with a [4f4] electron config-
uration were also carried out. With this traditional electronic
configuration, a 2.727 Å Nd−Cnt distance is predicted which
is much longer than the experimentally observed value of 2.563
Å. Also, no appreciable absorptions are predicted in the visible
spectrum, in contrast to the strong experimental absorbance.
E
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
with this view and qualitatively reproduce the much stronger
visible absorption of the La, Ce, Pr, Nd, Gd, and Tb
compounds compared to the Sm and Dy compounds; see
Figure 7.
are stable for at least 48 h. However, for Dy, the reduction
must be done at −35 °C in order to isolate a dark-colored
product. For Ho, Er, and Y, reductions, even at −35 °C, yield
unstable black solutions that do not persist longer than 12 h at
−35 °C. Only the dark-colored solutions of 3-La, 3-Ce, and 3-
Nd persist as long as the larger members of the 2-Ln series
with Ln = La−Gd. Hence, it appears that the 2-Ln complexes
of the larger lanthanides are more stable than the 3-Ln
analogues.
The thermal decomposition of 2-Tb was examined in a
manner analogous to previous studies on 3-Ln and [K-
(crypt)][Cp″3Ln]8,9 by monitoring the decrease in the
maximum absorbance in the UV−vis spectrum, in this case
at λmax = 784 nm. The data on 2-Tb were best modeled with
second order kinetics (see the Supporting Information), which
is consistent with previous Ln(II) decomposition studies.8,9
The observed second order rate constant of 5.4(7) × 10−2 M−1
cm−1 is an order of magnitude larger than the second order
rate constant, 4.25(5) × 10−3 M−1 cm−1, observed for 3-Tb.9
Hence, for this smaller metal, the 2-Tb complex is less stable
than the 3-Tb.
Figure 7. Predicted spectra from TDDFT using the TPSSh functional
and the f-in-core SCecp-mwb-d ECP (see the Supporting Information
for details).
DISCUSSION
■
Analysis of time-dependent density functional theory
(TDDFT) simulations for the (CptetLn)1− (Ln = La, Ce, Pr,
Nd, Gd, and Tb) compounds reveals that, in addition to the
MLCT absorptions found previously in the 3-Ln series, there
are strongly dipole-allowed Ln 5d → 6p and 5d → π*
transitions (Table S26, Supporting Information) at lower
energy. The three distinct bands in the computed visible
spectra can be associated with transitions to 6p Rydberg
orbitals of slightly different energies. The higher oscillator
strengths for the 5d → 6p transitions in the 2-Ln series can be
rationalized by the larger metal−ligand distances in these
compounds, which stabilize metal 6p Rydberg orbitals relative
to the 3-Ln compounds. MLCT transitions are still present in
2-Ln (Ln = La, Ce, Pr, Nd, Gd, and Tb), but they are much
weaker than these 5d → 6p transitions. It should also be noted
that 5d → 6p absorptions are also present in 3-Ln, but they are
weaker in those complexes.
Geometric effects may also be operative in the difference in
intensities of the transitions in 2-Ln vs 3-Ln. The C3h 2-Ln
complexes are more symmetrical when compared to the 3-Ln
species. In the 3-Ln series, two of the Me3Si substituents are
above the plane defined by the metal and the three Cp′ ring
centroids and one Me3Si group is below. In the 2-Ln series, the
arrangement of substituents has a different symmetry: the one
unique ring carbon atom is the C−H unit rather than the
unique ring substituted C−SiMe3 moiety in 3-Ln. In 2-Ln,
these unique C−H positions are more similar in the three rings
and are in or near the plane of the three ring centroids (see
Supporting Information Table S2). As a result, there are six
methyl groups above the plane of the three Cptet ring centroids
and six below. These differences in molecular geometry could
change the overlap of the occupied and virtual molecular
orbitals, which results in the difference in oscillator strengths
for the two types of transitions and the difference in extinction
coefficients in the spectra.
Although Cptet was originally considered to be an unlikely
ligand to stabilize 4fn5d1 Ln(II) ions because it is so electron
donating, it has provided crystallographically characterizable
tris(cyclopentadienyl) Ln(II) complexes for La, Ce, Pr, Nd,
Gd, and Tb, as well as the traditional Sm(II) and configura-
tional crossover Dy(II). The EPR data on [K(crypt)]-
[Cptet3La], 2-La, and the product of reducing Cptet3Y show
larger hyperfine coupling values consistent with more electron
density on the metal. Previously, reducing the electron density
on the metal was thought to be critical in isolating Ln(II)
complexes. Given the isolation of these 2-Ln complexes, it is
clear that cyclopentadienyl ligands with alkyl substituents are
not too electron donating to form Ln(II) complexes of this
type. The isolation of the 2-Ln complexes as well as a series of
[Ln(NR2)3]1− complexes40 show that using the hyperfine
coupling constant value for yttrium as a gauge for predicting
the stability of Ln(II) complexes is not appropriate! It remains
to be determined if this is a special aspect of yttrium and its 4d
orbitals.
There are two major differences between [K(crypt)]-
[Cptet3Ln], 2-Ln, and the previously reported [K(crypt)]-
[Cp′3Ln], 3-Ln, series. One notable difference for the six
4fn5d1 ions of La, Ce, Pr, Nd, Gd, and Tb is that the Ln−Cnt
distances for the Ln(II) Cptet complexes, 2-Ln, are 0.043−
0.048 Å longer than those of the analogous Ln(II) Cp′
complexes, 3-Ln. Hence, although the Cptet ligands may be
more electron donating, they are further away from the metal.
This may mitigate their greater electron donating ability and
allow the isolation of Ln(II) complexes for these metals. Of
course, the Cptet ligands also provide more steric protection for
the metal with their four substituents compared to the
monosubstituted Cp′. The importance of steric saturation to
the stability of Ln(III) complexes is well established.
It is interesting to note that the difference in metal−Cnt
distances between 2-Ln and 3-Ln for the traditional Sm(II)
complexes is only 0.022, whereas the difference for the
configurational crossover Dy(II) complex is 0.10 Å. Hence,
these 4fn+1 systems do not show regularity in the difference in
metal−ring centroid distances between Cptet and Cp′ ligand
systems.
Thermal Stability. The stability of the 2-Ln complexes of
the larger metals, Ln = La, Ce, Pr, Nd, Sm, and Gd, is greater
than that of Ln = Tb, Dy, Ho, Er, and Y. For the larger metals,
the dark blue/black solutions obtained by reduction with KC8
in THF, followed by Et2O layer diffusion at room temperature,
F
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 8. Plot of the 4fn+1 to 4fn5d1 promotion energy compared to the change in distance upon reduction of Cpx3Ln to (Cpx3Ln)1− for Cpx = Cptet
(left) and Cpx = Cp′ (right).
Another notable difference between 2-Ln and 3-Ln involves
the UV−vis spectra. The absorptions with the largest λmax
values are at lower energies compared to the most intense
absorptions of 3-Ln for Ln = La, Ce, Pr, Nd, Gd, and Tb.
TDDFT studies indicate that the maximum absorbances for
the two series do not arise from analogous transitions. For 2-
Ln, the most intense absorptions arise from Ln 5d → 6p
transitions; for 3-Ln, the largest absorptions are attributed to
MLCT. While both transitions are present in both complexes,
the MLCT transitions are more intense for 3-Ln and the
metal−metal transitions are more intense for 2-Ln. It is likely
that the longer metal−ligand distances in 2-Ln weaken the
MLCT and enhance the 5d → 6p transitions. The difference in
intensities of the two series could also occur due to the
difference in the geometry of the complexes. In addition to the
changes in metal−ligand distance, the higher molecular
symmetry of the 2-Ln complexes could be the cause of greater
metal p-character in lower energy unoccupied MOs, which
would have a stronger oscillator strength due to a large 5d →
6p transition dipole moment.
Ln complexes which have reduction potentials less negative
than −2.9 V.
CONCLUSION
■
The isolation of (Cptet3Ln)1− complexes for Ln = La, Ce, Pr,
Nd, Gd, and Tb indicates that the prior assumption that
electron-rich cyclopentadienyl ligands cannot be used to isolate
4fn5d1 Ln(II) complexes is incorrect. More generally, this
suggests that the new mixed configuration Ln(II) ions are
likely to be accessible with a variety of ligand systems. This will
allow a broader investigation of both their physical properties
and their reactivity. In addition, the results show the
importance of the ligand in determining which configuration,
4fn+1 or 4fn5d1, is adopted by Ln(II) ions in tris-
(cyclopentadienyl) coordination environments. The crossover
point between the two configurations is variable depending on
the ligands. Thus, judicious choice of the ligand environment
may profoundly affect the chemical, optical, and magnetic
properties of divalent lanthanides which can be significantly
different for 4fn+1 vs 4fn5d1 configurations.
In the [K(crypt)][Cptet3Ln], 2-Ln, series, both the crystallo-
graphic data and the UV−vis data suggest that 4fn5d1 ions are
present for La Ce, Pr, Nd, Gd, and Tb and 4fn+1 ions for Sm
and Dy. This is supported by the DFT calculations (see the
Supporting Information). This dichotomy differs from that
with the [K(crypt)][Cp′3Ln], 3-Ln, series where the Dy
complex was assigned a 4fn5d1 electron configuration. This is
consistent with the realization that the ligand field can affect
the configurations involving the 5d orbitals and demonstrates
that changing the substitution on the cyclopentadienyl rings
can change the border between 4fn5d1 and 4fn+1 configurations.
This is shown graphically in Figure 8 where the differences in
Ln−Cnt distances for Ln(II) and Ln(III) analogues are plotted
versus the 4fn+1 to 4fn5d1 promotion energy determined from
gas phase atomic spectra of +2 ions.57 The dotted line border
between traditional and nontraditional electron configurations
has changed for Cptet compared to Cp′. It is interesting to note
that attempts to make 2-Ho and 2-Er in this study led to
solutions too reactive to isolate. According to Figure 5, which
is admittedly a simplistic analysis of the system, 2-Ho and 2-Er
EXPERIMENTAL DETAILS
■
All manipulations and syntheses described below were conducted
with rigorous exclusion of air and water using standard Schlenk line
and glovebox techniques under an argon atmosphere. Solvents were
sparged with UHP argon (Airgas) and dried by passage through
columns containing Q-5 and molecular sieves prior to use. THF-d8
(Cambridge Isotope Laboratories) was dried over NaK alloy,
degassed by three freeze−pump−thaw cycles, and vacuum transferred
58
before use. KC8 was prepared as previously described. KCptet (Cptet
= C5Me4H) and Cptet3Ln (Ln = Y,59 La,45 Ce,47 Pr,47 Nd,43 Sm,45
Tb45), 1-Ln, were synthesized using modified literature procedures.
2.2.2-Cryptand (4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo[8,8,8]-
hexacosane, VWR) was held under a vacuum (1 mTorr) for 12 h
prior to use. The Evans method magnetic susceptibility51−53 was
obtained on a CRYO500 MHz spectrometer at 298 K unless
otherwise stated and referenced internally to residual protio-solvent
resonances. UV−vis measurements were conducted at 298 K in THF
with a Jasco 670 UV−vis-NIR spectrophotometer. IR samples were
prepared as KBr pellets on a Jasco FT/IR-4700 spectrometer.
Elemental analyses were conducted on a PerkinElmer 2400 Series II
CHNS elemental analyzer.
Synthesis of (C5Me4H)3Gd, 1-Gd. After GdCl3 (0.110 g, 0.418
mmol) was stirred in THF (10 mL) for 2 h, KCptet (0.201 g, 1.257
mmol) was added and the mixture was stirred for 12 h. The solvent
was removed, toluene (10 mL) was added, and the mixture has heated
to reflux in a sealed flask under a slight vacuum (∼85 °C) for 12 h.
could be 4fn+1 ions in the (Cptet 3− coordination environment.
)
3
Calculations of 4fn Ln(III) to 4fn+1 Ln(II) reduction potentials
for Ho and Er are −2.9 and −3.1 V, respectively; i.e., they
could be more reactive than Ho(II) and Er(II) ions in the 3-
G
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
The mixture was centrifuged, and the supernatant was filtered to give
a pale-yellow solution. Removal of the solvent (in vacuo) followed by
trituration with hexane produced a pale yellow solid, 1-Gd (0.111 g,
51%). Orange single crystals suitable for X-ray diffraction were
obtained from a concentrated solution of toluene. IR (cm−1): 3113m,
2962vs, 2905vs, 2885vs, 2722s, 2499w, 2046w, 1645w, 1580s, 1544s,
1484s, 1436vs, 1381vs, 1371vs, 1329s, 1173s, 1133m, 1107m, 1018s,
971s, 776vs, 699m, 610s. Anal. Calcd for C27H39Gd: C, 62.26; H,
mmol) to form a black crystalline solid of 2-Nd (0.033 g, 45%). Single
crystals of 2-Nd suitable for X-ray diffraction were obtained by layer
diffusion of the concentrated THF solution with Et2O at −35 °C. IR
(cm−1): 3080w, 2962m, 2883s, 2822s, 2704w, 1477m, 1444m,
1354m, 1323w, 1296m, 1258m, 1238w, 1173w, 1134s, 1109vs,
1081s, 951s, 933m, 832w, 749m, 700w. UV−visible (THF) λmax nm
(ε, M−1 cm−1): 833 (5600). Anal. Calcd for C45H75KN2O6Nd: C,
58.53; H, 8.19; N, 3.03. Found: C, 56.96; H, 7.96; N, 1.72. The low
values suggest incomplete sample combustion. The results give a
formula of C45H75N1.2.
7.55. Found: C, 59.87; H, 7.36. The results give a formula of C27H39.6
.
Obtaining satisfactory analysis on the new members of the known 1-
Ln series43,45,47,60,61 as well as the 2-Ln series has been unexpectedly
challenging, although problems with elemental analyses of organo-
lanthanides are well documented.7,62−64 Multiple attempts have been
made to get good data, but poor carbon combustion has been a
problem.
[K(crypt)][(C5Me4H)3Sm], 2-Sm. In a procedure analogous to the
preparation of 2-La, 1-Sm (0.029 g, 0.056 mmol) and crypt (0.021 g,
0.056 mmol) were dissolved in THF (2 mL) and reacted with KC8
(0.010 g, 0.074 mmol) to form a dark brown crystalline solid
identified as [K(crypt)][(C5Me4H)3Sm], 2-Sm (0.023 g, 43%). Single
dark brown crystals of 2-Sm suitable for X-ray diffraction were
obtained by layer diffusion of the concentrated THF solution with
Et2O at −35 °C. IR (cm−1): 3037w, 2911s, 2882vs, 2851s, 2817m,
2755w, 2717w, 1958w, 1477m, 1450m, 1444m, 1353s, 1264m,
1259m, 1237m, 1119s, 1106vs, 1028w, 949s, 932m, 877m, 818s,
754m, 657m. UV−visible (THF) λmax nm (ε, M−1 cm−1): 466 (200).
Anal. Calcd for C45H75KN2O6Sm: C, 58.15; H, 8.13; N, 3.01. Found:
C, 56.34; H, 8.01; N, 3.34. The results give a formula of C45H76N2.3.
[K(crypt)][(C5Me4H)3Gd], 2-Gd. In a procedure analogous to the
preparation of 2-La, 1-Gd (0.007 g, 0.01 mmol) and crypt (0.005 g,
0.01 mmol) were dissolved in THF (2 mL) and reacted with KC8
(0.003 g, 0.02 mmol) to form a dark blue crystalline solid identified as
[K(crypt)][(C5Me4H)3Gd], 2-Gd (0.011 g, 93%). Single dark blue
crystals of 2-Gd suitable for X-ray diffraction were obtained by layer
diffusion of the concentrated THF solution with Et2O at −35 °C. IR
(cm−1): 3087w, 2961m, 2880s, 2820s, 2702w, 1477m, 1445m, 1353s,
1293m, 1258m, 1118s, 1108vs, 1100s, 952s, 933m, 833w, 750m,
734m. UV−visible (THF) λmax nm (ε, M−1 cm−1): 745 (2500). Anal.
Calcd for C45H75KN2O6Gd: C, 57.72; H, 8.07; N, 2.99. Found: C,
56.40; H, 7.87; N, 2.92. The results give a formula of C45H74.8N2.
[K(crypt)][(C5Me4H)3Tb], 2-Tb. In a procedure analogous to the
preparation of 2-La, 1-Tb (0.032 g, 0.06 mmol) and crypt (0.028 g,
0.08 mmol) were dissolved in THF (2 mL) and reacted with KC8
(0.015 g, 0.11 mmol) to form a dark blue crystalline solid identified as
[K(crypt)][(C5Me4H)3Tb], 2-Tb (0.047 g, 83%). Single dark blue
crystals of 2-Tb suitable for X-ray diffraction were obtained by layer
diffusion of the concentrated THF solution with Et2O at −35 °C. IR
(cm−1): 3089w, 2961sm, 2882s, 2818s, 2701m, 1478m, 1444m,
1351m, 1295m, 1257m, 1134s, 1108vs, 1080s, 1078m, 952s, 933m,
832w, 822w, 749m, 734m, 603w, 567w, 525w. UV−visible (THF)
λmax nm (ε, M−1 cm−1): 784 (650). Anal. Calcd for C45H75KN2O6Tb:
C, 57.61; H, 8.06; N, 2.99. Found: C, 51.76; H, 6.83; N, 3.01. The
results give a formula of C45H70.8N2.2.
Synthesis of (C5Me4H)3Dy, 1-Dy. Similarly to 1-Gd, KCptet
(0.190 g, 1.185 mmol) was added stepwise to a solution of DyCl3
(0.100 g, 0.372 mmol) in THF (10 mL). After reflux and collection of
the resulting supernatant, a pale orange solid was obtained, 1-Dy
(0.1012 g, 95%). Orange single crystals suitable for X-ray diffraction
were obtained from a concentrated solution of toluene. IR (cm−1):
3117w, 2965s, 2901vs, 2856vs, 2723m, 1653m, 1436s, 1382s, 1327m,
1173m, 1141m, 1019s, 971s, 781vs, 668m, 608m. Anal. Calcd for
C27H39Dy: C, 61.64; H, 7.47. Found: C, 57.54; H, 7.23. The low
values suggest incomplete sample combustion. The results give a
formula of C27H40.4
.
Synthesis of [K(crypt)][(C5Me4H)3La], 2-La. 1-La (0.028 g,
0.056 mmol) and crypt (0.021 g, 0.057 mmol) were dissolved in THF
(2 mL). The solution was cooled to −35 °C prior to addition of KC8
(0.012 g, 0.085 mmol). The solution was immediately filtered, layered
using chilled hexane, and left at −35 °C. A dark blue crystalline solid,
identified as [K(crypt)][(C5Me4H)3La], 2-La, by X-ray diffraction,
was obtained after 3 days (0.046 g, 89%). Single dark blue crystals of
2-La suitable for X-ray diffraction were obtained by diffusion of the
concentrated THF solution with Et2O at −35 °C. IR (cm−1): 3070w,
2962m, 2882s, 2820s, 2704w, 1477m, 1445m, 1353s, 1321w, 1293m,
1258m, 1239w, 1173w, 1133s, 1108vs, 1097s, 1017w, 951s, 933m,
833w, 750m, 733w. UV−visible (THF) λmax nm (ε, M−1 cm−1): 745
(1700). Anal. Calcd for C45H75KN2O6La: C, 58.87; H, 8.23; N, 3.05.
Found: C, 58.81; H, 8.17; N, 3.36. The results give a formula of
C45H74.5N2.2.
[K(crypt)][(C5Me4H)3Ce], 2-Ce. In a procedure analogous to the
preparation of 2-La, 1-Ce (0.055 g, 0.11 mmol) and crypt (0.045 g,
0.12 mmol) were dissolved in THF (2 mL) and reacted with KC8
(0.022 g, 0.16 mmol) to form a dark blue crystalline solid identified as
[K(crypt)][(C5Me4H)3Ce], 2-Ce (0.112 g, 61%). Single dark blue
crystals of 2-Ce suitable for X-ray diffraction were obtained by layer
diffusion of the concentrated THF solution with Et2O at −35 °C. IR
(cm−1): 3074w, 2961m 2883s, 2820s, 2704w, 1477m, 1456m, 1444m,
1354m, 1295m, 1258m, 1134s, 1108vs, 1080s, 1018m, 951s, 933m,
833w, 749m, 734w. UV−visible (THF) λmax nm (ε, M−1 cm−1): 874
(4100). Anal. Calcd for C45H75KN2O6Ce: C, 58.79; H, 8.22; N, 3.05.
Found: C, 58.04; H, 8.34; N, 3.66. The results give a formula of
C45H77N2.4.
[K(crypt)][(C5Me4H)3Pr], 2-Pr. Synthesis of [K(crypt)]-
[(C5Me4H)3Pr], 2-Pr. Similar to the preparation of 2-Pr, 1-Pr
(0.015 g, 0.0274 mmol) and crypt (0.010 g, 0.0274 mmol) were
dissolved in THF (2 mL) and reacted with KC8 (0.005 mg, 0.03
mmol) to form a black crystalline solid of [K(crypt)][(C5Me4H)3Pr],
2-Pr (0.017 g, 76%). Single crystals of 2-Pr suitable for X-ray
diffraction were obtained by layer diffusion of the concentrated THF
solution with Et2O at −35 °C. IR (cm−1): 3077w, 2961m, 2882s,
2821s, 2702w, 1477m, 1445m, 1354m, 1296m, 1258m, 1134s,
1109vs, 1080s, 952s, 933m, 833w, 749m, 734w. UV−visible (THF)
λmax nm (ε, M−1 cm−1): 854 (4500). Anal. Calcd for C45H75KN2O6Pr:
C, 58.74; H, 8.22; N, 3.04. Found: C, 57.15; H, 7.88; N, 3.60. The
results give a formula of C45H74N2.4.
[K(crypt)][(C5Me4H)3Dy], 2-Dy. Similar to the preparation of 2-
La, 1-Dy (0.025 g, 0.05 mmol) and crypt (0.020 g, 0.05 mmol) were
dissolved in THF (2 mL) and reacted with KC8 (0.011 g, 0.08 mmol)
to form a dark brown crystalline solid identified as [K(crypt)]-
[(C5Me4H)3 Dy], 2-Dy (0.032 g, 68%). Single dark brown crystals of
2-Dy suitable for X-ray diffraction were obtained by layer diffusion of
the concentrated THF solution with Et2O at −35 °C. IR (cm−1):
3093w, 2955m, 2882vs, 2725w, 1475m, 1442m, 1353s, 1296m,
1258m, 1173w, 1118m, 1105vs, 1078s, 951s, 930m, 828w, 752m.
UV−visible (THF) λmax nm (ε, M−1 cm−1): 766 (200). Anal. Calcd
for C45H75KN2O6Dy: C, 57.40; H, 8.03; N, 2.97. Found: C, 56.34; H,
7.48; N, 3.08. The results give a formula of C45H71.2N2.1.
X-ray Crystallographic Data. Crystallographic information for
complexes 1-Ln (Ln = Dy, Gd) and 2-Ln (Ln = La, Ce, Pr, Nd, Sm
Gd, Tb, Dy) is summarized in Table 1 and in the Supporting
Information.
[K(crypt)][(C5Me4H)3Nd], 2-Nd. Similar to the preparation of 2-
La, 1-Nd (0.040 g, 0.079 mmol) and crypt (0.030 g, 0.079 mmol)
were dissolved in THF (2 mL) and reacted with KC8 (0.013 g, 0.095
H
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(7) Hitchcock, P. B.; Lappert, M. F.; Maron, L.; Protchenko, A. V.
Lanthanum does form stable molecular compounds in the + 2
oxidation state. Angew. Chem., Int. Ed. 2008, 47, 1488−91.
(8) Palumbo, C. T.; Darago, L. E.; Windorff, C. J.; Ziller, J. W.;
Evans, W. J. Trimethylsilyl versus Bis(trimethylsilyl) Substitution in
Tris(cyclopentadienyl) Complexes of La, Ce, and Pr: Comparison of
Structure, Magnetic Properties, and Reactivity. Organometallics 2018,
37, 900−905.
(9) MacDonald, M. R.; Bates, J. E.; Ziller, J. W.; Furche, F.; Evans,
W. J. Completing the series of + 2 ions for the lanthanide elements:
synthesis of molecular complexes of Pr2+, Gd2+, Tb2+, and Lu2+. J. Am.
Chem. Soc. 2013, 135, 9857−68.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.8b00557.
Details of crystallographic data collection, structure
solution, and refinement, DFT calculation metrical
data on 1-Ln (Gd, Dy), 2-Ln (La, Ce, Pr, Nd, Sm,
Gd, Tb, Dy), and 3-Ln (La, Ce, Pr, Nd, Sm, Gd, Tb,
Dy) (PDF)
(10) Fieser, M. E.; MacDonald, M. R.; Krull, B. T.; Bates, J. E.;
Ziller, J. W.; Furche, F.; Evans, W. J. Structural, spectroscopic, and
theoretical comparison of traditional vs recently discovered Ln2+ ions
in the [K(2.2.2-cryptand)][(C5H4SiMe3)3Ln] complexes: the variable
nature of Dy2+ and Nd2+. J. Am. Chem. Soc. 2015, 137, 369−82.
(11) MacDonald, M. R.; Ziller, J. W.; Evans, W. J. Synthesis of a
, [(18-crown-6)K]-
[(C5H4SiMe3)3Y]. J. Am. Chem. Soc. 2011, 133, 15914−7.
(12) MacDonald, M. R.; Bates, J. E.; Fieser, M. E.; Ziller, J. W.;
Furche, F.; Evans, W. J. Expanding Rare-Earth Oxidation State
Chemistry to Molecular Complexes of Holmium(II) and Erbium(II).
J. Am. Chem. Soc. 2012, 134, 8420−8423.
(13) Langeslay, R. R.; Fieser, M. E.; Ziller, J. W.; Furche, F.; Evans,
W. J. Synthesis, structure, and reactivity of crystalline molecular
complexes of the {[C5H3(SiMe3)2]3Th}1‑ anion containing thorium in
the formal + 2 oxidation state. Chemical Science 2015, 6, 517−521.
(14) MacDonald, M. R.; Fieser, M. E.; Bates, J. E.; Ziller, J. W.;
Furche, F.; Evans, W. J. Identification of the + 2 Oxidation State for
Uranium in a Crystalline Molecular Complex, [K(2.2.2-Cryptand)]-
[(C5H4SiMe3)3U]. J. Am. Chem. Soc. 2013, 135, 13310−13313.
(15) Windorff, C. J.; MacDonald, M. R.; Meihaus, K. R.; Ziller, J.
W.; Long, J. R.; Evans, W. J. Expanding the Chemistry of Molecular
U2+ Complexes: Synthesis, Characterization, and Reactivity of the
{[C5H3(SiMe3)2]3U}− Anion. Chem. - Eur. J. 2016, 22, 772−82.
(16) Windorff, C. J.; Chen, G. P.; Cross, J. N.; Evans, W. J.; Furche,
F.; Gaunt, A. J.; Janicke, M. T.; Kozimor, S. A.; Scott, B. L.
Identification of the Formal + 2 Oxidation State of Plutonium:
Synthesis and Characterization of {Pu(II)[C5H3(SiMe3)2]3}−. J. Am.
Chem. Soc. 2017, 139, 3970−3973.
Accession Codes
CCDC 1857412−1857421 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
crystalline molecular complex of Y2+
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: filipp.furche@uci.edu.
*E-mail: wevans@uci.edu.
ORCID
Tener F. Jenkins: 0000-0002-9469-7405
David H. Woen: 0000-0002-5764-1453
Luke N. Mohanam: 0000-0002-7862-5153
Joseph W. Ziller: 0000-0001-7404-950X
Filipp Furche: 0000-0001-8520-3971
William J. Evans: 0000-0002-0651-418X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the U.S. National Science Foundation for support of
the experimental studies (CHE-1565776 to W.J.E.) and the
theoretical studies (CHE-1800431 to F.F.). For crystallo-
graphic assistance, we thank Michael Wojnar and Austin J.
Ryan. We also thank Professor A. S. Borovik and the Laser
Spectroscopy Facility for spectroscopic assistance.
(17) Su, J.; Windorff, C. J.; Batista, E. R.; Evans, W. J.; Gaunt, A. J.;
Janicke, M. T.; Kozimor, S. A.; Scott, B. L.; Woen, D. H.; Yang, P.
Identification of the Formal + 2 Oxidation State of Neptunium:
S y n t h e s i s a n d S t r u c t u r a l C h a r a c t e r i z a t i o n o f
{NpII[C5H3(SiMe3)2]3}1−. J. Am. Chem. Soc. 2018, 140, 7425−7428.
(18) Gun’ko, Y. K.; Hitchcock, P. B.; Lappert, M. F. Activation of a
C−O bond by reaction of a tris(cyclopentadienyl)lanthanide complex
with an alkali metal in dimethoxyethane (DME); crystal structures of
[Nd{η-C5H3(SiMe3)2-1,3}2(μ-OMe)2Li(DME)] and [{Ce(η-
REFERENCES
■
(1) Evans, W. J. Tutorial on the Role of Cyclopentadienyl Ligands in
the Discovery of Molecular Complexes of the Rare-Earth and Actinide
Metals in New Oxidation States. Organometallics 2016, 35, 3088−
3100.
t
C5H3 Bu2-1,3)2(μ-OMe)2]. J. Organomet. Chem. 1995, 499, 213−219.
(19) Cassani, M. C.; Duncalf, D. J.; Lappert, M. F. The First
Example of a Crystalline Subvalent Organolanthanum Complex:
[K([18]crown-6)- (η2-C6H6)2][(LaCptt2)2(μ-η6:η6-C6H6)]
•2C6H6(Cptt= η5-C5H3But2-1,3). J. Am. Chem. Soc. 1998, 120,
12958−12959.
(2) Woen, D. H.; Evans, W. J. Expanding the + 2 Oxidation State of
the Rare-Earth Metals, Uranium, and Thorium in Molecular
Complexes. In Handbook on the Physics and Chemistry of Rare Earths;
(20) Jaroschik, F.; Nief, F.; Le Goff, X. F.; Ricard, L. Isolation of
stable organodysprosium(II) complexes by chemical reduction of
dysprosium(III) precursors. Organometallics 2007, 26, 1123−1125.
(21) Jaroschik, F.; Momin, A.; Nief, F.; Le Goff, X.-F.; Deacon, G.
B.; Junk, P. C. Dinitrogen Reduction and C−H Activation by the
Bunzli, J.-C. G., Pecharsky, V. K., Eds.; Elsevier: Amsterdam, The
̈
Netherlands, 2016; Vol. 50, pp 337−394.
(3) Meyer, G. Reduced Halides of the Rare-Earth Elements. Chem.
Rev. 1988, 88, 93−107.
(4) Meyer, G. The Divalent State in Solid Rare-Earth Metal Halides;
John Wiley & Sons: Chichester, U.K., 2012.
t
Divalent Organoneodymium Complex [(C5H2 Bu3)2Nd(μ-I)K([18]-
crown-6)]. Angew. Chem. 2009, 121, 1137−1141.
(5) Bochkarev, M. N. Molecular compounds of “new” divalent
lanthanides. Coord. Chem. Rev. 2004, 248, 835−851.
(6) Nief, F. Molecular chemistry of the rare-earth elements in
uncommon low-valent states. In Handbook on the Physics and
̈
Chemistry of Rare Earths; Gschneidner, K. A., Jr.; Bunzli, J.-C. G.;
Pecharsky, V. K., Eds.; Elsevier Science: Amsterdam, The Nether-
(22) La Pierre, H. S.; Scheurer, A.; Heinemann, F. W.; Hieringer,
W.; Meyer, K. Synthesis and characterization of a uranium(II)
monoarene complex supported by delta backbonding. Angew. Chem.,
Int. Ed. 2014, 53, 7158−62.
(23) Fieser, M. E.; Ferrier, M. G.; Su, J.; Batista, E.; Cary, S. K.;
Engle, J. W.; Evans, W. J.; Lezama Pacheco, J. S.; Kozimor, S. A.;
lands, 2010; Vol. 40, pp 241−300.
I
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Olson, A. C.; Ryan, A. J.; Stein, B. W.; Wagner, G. L.; Woen, D. H.;
Vitova, T.; Yang, P. Evaluating the electronic structure of formal LnII
(40) Ryan, A. J.; Darago, L. E.; Balasubramani, S. G.; Chen, G. P.;
Ziller, J.; Furche, F.; Evans, W. J. Synthesis, Structure, and Magnetism
of Tris(amide) [Ln{N(SiMe3)2}3]1− Complexes of the Non-tradi-
tional + 2 Lanthanide Ions. Chem. - Eur. J. 2018, 24, 7702−7709.
(41) In order to support this assignment, the electride species was
independently prepared from the reaction of crypt and KC8 in THF.
The EPR spectrum of that product gave a signal at g = 2.001, Figure
S1. This extra feature was still present even when less than 1 equiv of
crypt was used. Consistent with this assignment, the signal became
more prominent in the room temperature spectrum possibly because
the electride species was more stable than the Y(II) product at this
temperature and only a small amount of the Y(II) species survived by
the time the spectrum was collected.
1‑
ions in LnII(C5H4SiMe3)3 using XANES spectroscopy and DFT
calculations. Chemical Science 2017, 8, 6076−6091.
(24) Meihaus, K. R.; Fieser, M. E.; Corbey, J. F.; Evans, W. J.; Long,
J. R. Record High Single-Ion Magnetic Moments Through 4fn5d1
Electron Configurations in the Divalent Lanthanide Complexes
[(C5H4SiMe3)3Ln]−. J. Am. Chem. Soc. 2015, 137, 9855−9860.
(25) Corbey, J. F.; Woen, D. H.; Palumbo, C. T.; Fieser, M. E.;
Ziller, J. W.; Furche, F.; Evans, W. J. Ligand Effects in the Synthesis of
Ln2+ Complexes by Reduction of Tris(cyclopentadienyl) Precursors
Including C−H Bond Activation of an Indenyl Anion. Organometallics
2015, 34, 3909−3921.
(26) Huh, D. N.; Kotyk, C. M.; Gembicky, M.; Rheingold, A. L.;
Ziller, J. W.; Evans, W. J. Synthesis of rare-earth-metal-in-cryptand
dications, [Ln(2.2.2-cryptand)]2+, from Sm2+, Eu2+, and Yb2+ silyl
metallocenes (C5H4SiMe3)2Ln(THF)2. Chem. Commun. 2017, 53,
8664−8666.
(42) Shannon, R. Revised effective ionic radii and systematic studies
of interatomic distances in halides and chalcogenides. Acta
Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1976,
32, 751−767.
(43) Windorff, C. J.; Dumas, M. T.; Ziller, J. W.; Gaunt, A. J.;
Kozimor, S. A.; Evans, W. J. Small-Scale Metal-Based Syntheses of
Lanthanide Iodide, Amide, and Cyclopentadienyl Complexes as
Analogues for Transuranic Reactions. Inorg. Chem. 2017, 56, 11981−
11989.
(44) Woen, D. H.; Huh, D. N.; Ziller, J. W.; Evans, W. J. Reactivity
of Ln(II) Complexes Supported by (C5H4Me)1− Ligands with THF
and PhSiH3: Isolation of Ring-Opened, Bridging Alkoxyalkyl,
Hydride, and Silyl Products. Organometallics 2018, 37, 3055−3063.
(45) Schumann, H.; Glanz, M.; Hemling, H.; Ekkehard Hahn, F.
Metallorganische Verbindungen der Lanthanoide. 93 [1]. Tetrame-
thylcyclopentadienyl-Komplexe ausgewhlter 4f-Elemente. Z. Anorg.
Allg. Chem. 1995, 621, 341−345.
(46) Xie, Z. W.; Chui, K. L.; Liu, Z. X.; Xue, F.; Zhang, Z. Y.; Mak,
T. C. W.; Sun, J. Systematic studies on the reactions of lanthanide
trichlorides with Na[1,3-bis(trimethylsilyl)cyclopentadienyl]. Crystal
structures of [1,3-(Me3Si)2C5H3]3Ln (Ln = La, Nd, Gd, Dy). J.
Organomet. Chem. 1997, 549, 239−244.
(47) Evans, W. J.; Rego, D. B.; Ziller, J. W. Synthesis, structure, and
15N NMR studies of paramagnetic lanthanide complexes obtained by
reduction of dinitrogen. Inorg. Chem. 2006, 45, 10790−8.
(48) Stults, S. D.; Andersen, R. A.; Zalkin, A. Structural Studies on
Cyclopentadienyl Compounds of Trivalent Cerium - Tetrameric
(Me5 H4 )3 Ce and Monomeric (Me3 SiC5 H4 )3 Ce and
[(Me3Si)2C5H3]3Ce and Their Coordination Chemistry. Organo-
metallics 1990, 9, 115−122.
(49) Minasian, S. G.; Krinsky, J. L.; Rinehart, J. D.; Copping, R.;
Tyliszczak, T.; Janousch, M.; Shuh, D. K.; Arnold, J. A comparison of
4f vs 5f metal-metal bonds in (CpSiMe3)3M-ECp* (M = Nd, U; E =
Al, Ga; Cp* = C5Me5): synthesis, thermodynamics, magnetism, and
electronic structure. J. Am. Chem. Soc. 2009, 131, 13767−83.
(50) Lappert, M. F.; Singh, A.; Atwood, J. L.; Hunter, W. E. Use of
the bis(trimethylsilyl)cyclopentadienyl ligand for stabilising early (f0−
f3) lanthanocene chlorides; X-ray structure of [(Pr{η-
[C5H3(SiMe3)2]}2Cl)2] and of isoleptic scandium and ytterbium
complexes. J. Chem. Soc., Chem. Commun. 1981, 0, 1190−1191.
(51) Evans, D. F. 400. The determination of the paramagnetic
susceptibility of substances in solution by nuclear magnetic resonance.
J. Chem. Soc. 1959, 2003−2005.
(27) Lauher, J. W.; Hoffmann, R. Structure and chemistry of
bis(cyclopentadienyl)-MLn complexes. J. Am. Chem. Soc. 1976, 98,
1729−1742.
(28) Bursten, B. E.; Rhodes, L. F.; Strittmatter, R. J. Bonding in
Tris(y5-cyclopentadienyl) Actinide Complexes. 2. On the Ground
Electronic Configurations of “Base-Free” Cp,An Complexes (An =
Th, Pa, U, Np, Pu). J. Am. Chem. Soc. 1989, 111, 2756−2758.
(29) Bursten, B. E.; Rhodes, L. F.; Strittmatter, R. J. Bonding in
Tris(y5-cyclopentadienyl) Actinide Complexes. 3. Interaction of 7-
Neutral, 7-Acidic, and 7-Basic Ligands with (T5-C5H5)3U. J. Am.
Chem. Soc. 1989, 111, 2758−2766.
(30) Strittmatter, R. J.; Bursten, B. E. Bonding in Tris(η5-
Cyclopentadienyl) Actinide Complexes 0.5. A Comparison of the
Bonding in Np, Pu, and Transplutonium Compounds with That in
Lanthanide Compounds and a Transition-Metal Analog. J. Am. Chem.
Soc. 1991, 113, 552−559.
(31) Lukens, W. W.; Andersen, R. A. Synthesis, Structure, and
Reactions of (η5-C5H5)3Zr. Organometallics 1995, 14, 3435−3439.
(32) Denning, R. G.; Harmer, J.; Green, J. C.; Irwin, M. Covalency
in the 4f Shell of tris-Cyclopentadienyl Ytterbium (YbCp3)A
Spectroscopic Evaluation. J. Am. Chem. Soc. 2011, 133, 20644−20660.
(33) Beck, H. P. Notizen: NdI2-II, eine metallisch leitende
Hochdruckmodifikation ?/NdI2, a Metallic High Pressure Modifica-
tion ? Z. Naturforsch., B: J. Chem. Sci. 1976, 31, 1548.
(34) Gerlitzki, N.; Meyer, G.; Mudring, A.-V.; Corbett, J. D.
Praseodymium diiodide, PrI2, revisited by synthesis, structure
determination and theory. J. Alloys Compd. 2004, 380, 211−218.
(35) Fang, M.; Lee, D. S.; Ziller, J. W.; Doedens, R. J.; Bates, J. E.;
Furche, F.; Evans, W. J. Synthesis of the (N2)3‑ radical from Y2+ and
its protonolysis reactivity to form (N2H2)2‑ via the Y[N(SiMe3)2]3/
KC8 reduction system. J. Am. Chem. Soc. 2011, 133, 3784−7.
(36) Zachmanoglou, C. E.; Docrat, A.; Bridgewater, B. M.; Parkin,
G.; Brandow, C. G.; Bercaw, J. E.; Jardine, C. N.; Lyall, M.; Green, J.
C.; Keister, J. B. The electronic influence of ring substituents and ansa
bridges in zirconocene complexes as probed by infrared spectroscopic,
electrochemical, and computational studies. J. Am. Chem. Soc. 2002,
124, 9525−46.
(37) Coles, M. P.; Hitchcock, P. B.; Lappert, M. F.; Protchenko, A.
V. Syntheses and Structures of the Crystalline, Highly Crowded 1,3-
Bis(trimethylsilyl)cyclopentadienyls [MCp″3] (M = Y, Er, Yb),
[PbCp″2], [{YCp″2(μ-OH)}2], [(ScCp″2)2(μ-η2:η2-C2H4)],
[YbCp″2Cl(μ-Cl)K(18-crown-6)], and [{KCp″}∞]. Organometallics
2012, 31, 2682−2690.
(38) Peterson, J. K.; MacDonald, M. R.; Ziller, J. W.; Evans, W. J.
Synthetic Aspects of (C5H4SiMe3)3Ln Rare-Earth Chemistry:
Formation of (C5H4SiMe3)3Lu via [(C5H4SiMe3)2Ln]+ Metallocene
Precursors. Organometallics 2013, 32, 2625−2631.
(52) Becconsall, J. K. Effect of magnetically anisotropic solvent
molecules on nuclear screening. Mol. Phys. 1968, 15, 129−139.
(53) Schubert, E. M. Utilizing the Evans method with a
superconducting NMR spectrometer in the undergraduate laboratory.
J. Chem. Educ. 1992, 69, 62.
(54) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence,
triple zeta valence and quadruple zeta valence quality for H to Rn:
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(55) Dolg, M.; Stoll, H.; Savin, A.; Preuss, H. Energy-adjusted
pseudopotentials for the rare earth elements. Theoretica chimica acta
1989, 75, 173−194.
(39) Woen, D. H.; Chen, G. P.; Ziller, J. W.; Boyle, T. J.; Furche, F.;
Evans, W. J. Solution Synthesis, Structure, and CO2 Reduction
Reactivity of a Scandium(II) Complex, {Sc[N(SiMe3)2]3}1−. Angew.
Chem., Int. Ed. 2017, 56, 2050−2053.
J
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(56) Gulde, R.; Pollak, P.; Weigend, F. Error-Balanced Segmented
Contracted Basis Sets of Double-ζ to Quadruple-ζ Valence Quality
for the Lanthanides. J. Chem. Theory Comput. 2012, 8, 4062−4068.
(57) Dorenbos, P. f → d transition energies of divalent lanthanides
in inorganic compounds. J. Phys.: Condens. Matter 2003, 15, 575.
(58) Bergbreiter, D. E.; Killough, J. M. Reactions of potassium-
graphite. J. Am. Chem. Soc. 1978, 100, 2126−2134.
̈
(59) Soller, B. S.; Sun, Q.; Salzinger, S.; Jandl, C.; Pothig, A.; Rieger,
B. Ligand Induced Steric Crowding in Rare Earth Metal-Mediated
Group Transfer Polymerization of Vinylphosphonates: Does Enthalpy
Matter? Macromolecules 2016, 49, 1582−1589.
(60) Shestakov, B. G.; Mahrova, T. V.; Larionova, J.; Long, J.;
Cherkasov, A. V.; Fukin, G. K.; Lyssenko, K. A.; Scherer, W.; Hauf,
C.; Magdesieva, T. V.; Levitskiy, O. A.; Trifonov, A. A. Ytterbium(III)
Complexes Coordinated by Dianionic 1,4-Diazabutadiene Ligands.
Organometallics 2015, 34, 1177−1185.
(61) Evans, W. J.; Lee, D. S.; Johnston, M. A.; Ziller, J. W. The
Elusive (C5Me4H)3Lu: Its Synthesis and LnZ3/K/N2 Reactivity.
Organometallics 2005, 24, 6393−6397.
(62) Goodwin, C. A. P.; Joslin, K. C.; Lockyer, S. J.; Formanuik, A.;
Morris, G. A.; Ortu, F.; Vitorica-Yrezabal, I. J.; Mills, D. P.
Homoleptic Trigonal Planar Lanthanide Complexes Stabilized by
Superbulky Silylamide Ligands. Organometallics 2015, 34, 2314−
2325.
(63) Chilton, N. F.; Goodwin, C. A. P.; Mills, D. P.; Winpenny, R. E.
P. The first near-linear bis(amide) f-block complex: a blueprint for a
high temperature single molecule magnet. Chem. Commun. 2015, 51,
101−103.
(64) Goodwin, C. A. P.; Chilton, N. F.; Vettese, G. F.; Moreno
Pineda, E.; Crowe, I. F.; Ziller, J. W.; Winpenny, R. E. P.; Evans, W. J.;
Mills, D. P. Physicochemical Properties of Near-Linear Lanthanide-
(II) Bis(silylamide) Complexes (Ln = Sm, Eu, Tm, Yb). Inorg. Chem.
2016, 55, 10057−10067.
K
DOI: 10.1021/acs.organomet.8b00557
Organometallics XXXX, XXX, XXX−XXX