Unsolvated lanthanide metallocene cations [(C5Me5)2Ln][BPh4]: Multiple syntheses, structural characterization, and reactivity including the formation of (C5Me5)3Nd

Article abstract of DOI:10.1021/ja980534o

Divalent (C₅Me₅)₂Sm reacts with AgBPh₄ in toluene to form [(C₅Me₅)₂Sm][BPh₄], 1, in ca. 60% yield. The solid-state structure of 1 consists of a trivalent (C₅Me₅)₂Sm bent metallocene unit with a 2.702(3) ? average Sm-C(C₅Me₅) distance that is oriented toward two of the phenyl rings of the [BPh₄]^- anion with 2.825(3) and 2.917(3) ? Sm-C(o-Ph) distances. 1 can also be obtained from reactions of Et₃NHBPh₄ in arene solvents with the trivalent samarium precursors (C₅Me₅)₂Sm[CH(SiMe₃)₂] (>50% yield) and (C₅Me₅)₂Sm(η³-CH₂CHC₂) (2) (>95% yield). 1 reacts with LiCH(SiMe₃)₂ in benzene to produce (C₅Me₅)₂Sm[CH(SiMe₃)₂] in over 95% yield. The reaction of 1 with KC₅Me₅ in benzene constitutes a new synthesis of the sterically crowded complex (C₅Me₅)₃Sm, which is formed in over 90% yield. This reaction provides a convenient way to make (C₅Me₅)₃Ln complexes with lanthanides which do not have a reactive divalent oxidation state. To enhance the ease of preparing (C₅Me₅)₃Ln complexes from LnCl₃, an improved synthesis of the allyl precursors (C₅Me₅)₂Ln(η³-CH₂CHCH₂) (Ln = Sm (2), Nd (3), Tm (4)) is reported. 2-4 can be prepared in 60-90% yield from (C₅Me₅)₂LnCl₂K(THF)₂ and ClMg(CH₂CHCH₂) followed by desolvation of the solids between 55 and 70°C for 4-16 h. 2-4 react with Et₃NHBP₄ in benzene to produce [(C₅Me₅)₂Ln][BPh₄] (Ln = Sm (1), Nd (5), Tm(6)). 5 has a solid-state structure identical to that of 1 and similarly reacts with LiCH(SiMe₃)₂ and KC₅Me₅ in benzene to produce (C₅Me₅)₂Nd[CH(SiMe₃)₂ and (C₅Me₅)₃Nd (7), respectively, in high yield. 7 was characterized by X-ray crystallography and shown to have an (η⁵-C₅Me₅)₃Nd structure with a 2.86(6) ? Nd-C(C₅Me₅) distance. Since the allyl complexes (C₅Me₅)₂Ln(η³-CH₂CHCH₂) are readily converted to the hydrides [(C₅Me₅)₂LnH](n) by hydrogen, the improved synthesis of the allyl complexes also provides an improved route to these hydrides as demonstrated by the reaction of (C₅Me₅)₂Nd(η³-CH₂CHCH₂) with H₂ to form [(C₅Me₅)₂NdH]₂ in 75% yield.

Full text of DOI:10.1021/ja980534o

J. Am. Chem. Soc. 1998, 120, 6745-6752
6745
Unsolvated Lanthanide Metallocene Cations [(C5Me5)2Ln][BPh4]:
Multiple Syntheses, Structural Characterization, and Reactivity
Including the Formation of (C5Me5)3Nd1
William J. Evans,* Christopher A. Seibel, and Joseph W. Ziller
Contribution from the Department of Chemistry, UniVersity of California, IrVine,
IrVine, California 92697-2025
ReceiVed February 17, 1998
Abstract: Divalent (C5Me5)2Sm reacts with AgBPh4 in toluene to form [(C5Me5)2Sm][BPh4], 1, in ca. 60%
yield. The solid-state structure of 1 consists of a trivalent (C5Me5)2Sm bent metallocene unit with a 2.702(3)
Å average Sm-C(C5Me5) distance that is oriented toward two of the phenyl rings of the [BPh4]- anion with
2.825(3) and 2.917(3) Å Sm-C(o-Ph) distances. 1 can also be obtained from reactions of Et3NHBPh4 in
arene solvents with the trivalent samarium precursors (C5Me5)2Sm[CH(SiMe3)2] (>50% yield) and (C5Me5)2-
Sm(η3-CH2CHCH2) (2) (>95% yield). 1 reacts with LiCH(SiMe3)2 in benzene to produce (C5Me5)2Sm[CH-
(SiMe3)2] in over 95% yield. The reaction of 1 with KC5Me5 in benzene constitutes a new synthesis of the
sterically crowded complex (C5Me5)3Sm, which is formed in over 90% yield. This reaction provides a
convenient way to make (C5Me5)3Ln complexes with lanthanides which do not have a reactive divalent oxidation
state. To enhance the ease of preparing (C5Me5)3Ln complexes from LnCl3, an improved synthesis of the
allyl precursors (C5Me5)2Ln(η3-CH2CHCH2) (Ln ) Sm (2), Nd (3), Tm (4)) is reported. 2-4 can be prepared
in 60-90% yield from (C5Me5)2LnCl2K(THF)2 and ClMg(CH2CHCH2) followed by desolvation of the solids
between 55 and 70 °C for 4-16 h. 2-4 react with Et3NHBPh4 in benzene to produce [(C5Me5)2Ln][BPh4]
(Ln ) Sm (1), Nd (5), Tm (6)). 5 has a solid-state structure identical to that of 1 and similarly reacts with
LiCH(SiMe3)2 and KC5Me5 in benzene to produce (C5Me5)2Nd[CH(SiMe3)2] and (C5Me5)3Nd (7), respectively,
in high yield. 7 was characterized by X-ray crystallography and shown to have an (η5-C5Me5)3Nd structure
with a 2.86(6) Å Nd-C(C5Me5) distance. Since the allyl complexes (C5Me5)2Ln(η3-CH2CHCH2) are readily
converted to the hydrides [(C5Me5)2LnH]n by hydrogen, the improved synthesis of the allyl complexes also
provides an improved route to these hydrides as demonstrated by the reaction of (C5Me5)2Nd(η3-CH2CHCH2)
with H2 to form [(C5Me5)2NdH]2 in 75% yield.
Introduction
to date. In organosamarium chemistry, the first metallocene
cation to be crystallographically characterized was [(C5Me5)2-
Sm(THF)2][BPh4].6 This complex can be conveniently made
Recent research in metallocene chemistry has shown that
metallocene cations can play an important role in organometallic
chemistry.2-5 The enhanced electrophilicity of a metallocene
bearing a formal positive charge can have an important influence
on reactivity. This is especially important in olefin polymeriza-
tions and transformations.2-5
7
by oxidizing the divalent precursor (C5Me5)2Sm(THF)2 with
(3) (a) Yi, C. S.; Wo´dka, D.; Rheingold A. L.; Yap, G. P. A.
Organometallics 1996, 15, 2. (b) Jordan, R. F.; Wo, Z. J. Am. Chem. Soc.
1995, 117, 5867. (c) Jordan, R. F.; Rodewald, S. J. Am. Chem. Soc. 1994,
116, 4491. (d) Jordan, R. F.; Alelyunas, Y. W.; Baenziger, N. C.; Bradley,
P. K. Organometallics 1994, 13, 148. (e) Stryker, J. M.; Tjaden, E. B.;
Casty, G. L. J. Am. Chem. Soc. 1993, 115, 9814. (f) Jordan, R. F.; Guram,
A. S.; Guo, Z. J. Am. Chem. Soc. 1993, 115, 4902. (g) Collins, S.; Hong,
Y.; Kuntz, B. A. Organometallics 1993, 12, 964. (h) Jordan, R. F.; Guram,
A. S. J. Org. Chem. 1993, 58, 5595. (i) Collins, S.; Hong, Y.; Norris, D. J.
J. Org. Chem. 1993, 58, 3591. (j) Jordan, R. F.; Guram, A. S.; Swenson D.
C. J. Am. Chem. Soc. 1992, 114, 8991. (k) Jordan, R. F.; Gurma, A. S. J.
Org. Chem. 1992, 57, 5994. (l) Jordan, R. F.; Gurma, A. S.; Taylor, D. F.
J. Am. Chem. Soc. 1991, 113, 1833. (m) Jordan, R. F.; Guram, A. S.
Organometallics 1991, 10, 3470. (n) Jordan, R. F.; Alelyunas, Y. W.; Echols,
S. F.; Borkowsky, S. L.; Bradley, P. K. Organometallics 1991, 10, 1406.
(o) Jordan, R. F.; Borkowsky, S. L.; Hinch, G. D. Organometallics 1991,
10, 1268. (p) Jordan, R. F.; Crowther, D. J.; Baenziger, N. C.; Verma, A.
Organometallics 1990, 9, 2574. (q) Jordan, R. F.; Guram, A. S. Organo-
metallics 1990, 9, 2190. (r) Jordan, R. F.; Guram, A. S. Organometallics
1990, 9, 2116. (s) Jordan, R. F.; Taylor, D. F.; Baenziger, N. C.
Organometallics 1990, 9, 1546. (t) Jordan, R. F.; Taylor, D. F. J. Am. Chem.
Soc. 1989, 111, 778. (u) Jordan, R. F.; Dasher, W. E.; Echols, S. F. J. Am.
Chem. Soc. 1986, 108, 1718.
Despite the importance of metallocene cations, relatively few
examples of lanthanide metallocene cations have been reported
(1) Reported in part at the 214th National Meeting of the American
Chemical Society, September 1997, Las Vegas, NV; Abstract 336.
(2) (a) Jordan, R. F.; Kim, I.; Nishihara, Y. Organometallics 1997, 16,
3314. (b) Jordan, R. F.; Tsukahara, T.; Swenson, D. C. Organometallics
1997, 16, 3303. (c) Bochmann, M. J. Chem. Soc., Dalton Trans. 1996, 255.
(d) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.; Waymouth,
R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (e) Bergman, R. G.;
Arndtsen, B. A. Science 1995, 270, 1970. (f) Marks, T. J.; Jia, L.; Yang,
X.; Ishihara, A. Organometallics 1995, 14, 3135. (g) Jordan, R. F.; Guo,
Z.; Swenson, D. C. Organometallics 1994, 13, 1424. (h) Jordan, R. F.;
Borkowsky, S. L.; Baenziger, N. C. Organometallics 1993, 12, 486. (i)
Collins, S.; Ward, D. G. J. Am. Chem. Soc. 1992, 114, 5460. (j) Teuben, J.
H.; Eshuis, J. J.; Tan, Y. Y.; Meetsma, A. Organometallics 1992, 11, 362.
(k) Bochmann, M.; Lancaster, S. J. J. Organomet. Chem. 1992, 434, C1.
(l) Collins, S.; Koene, B. E.; Ramachandran, R.; Taylor, N. J. Organome-
tallics 1991, 10, 2092. (m) Bochmann, M.; Jaggar, A. J.; Nicholls, J. C.
Angew. Chem., Int. Ed. Engl. 1990, 29, 780. (n) Jordan, R. F.; Bradley, P.
K.; LaPointe, R. E.; Taylor, D. F. New J. Chem. 1990, 14, 505. (o) Jordan,
R. F.; LaPointe, R. E.; Bradley, P. K.; Baenziger, N. Organometallics 1989,
8, 2992. (p) Jordan, R. F.; Bajgur, C. S.; Willett, R.; Scott, B. J. Am. Chem.
Soc. 1986, 108, 7410.
(4) (a) Brookhart, M.; Hauptman, E. Organometallics 1994, 13, 774.
(b) Brookhart, M.; Hauptman, E.; Lincoln, D. M. J. Am. Chem. Soc. 1992,
114, 10394. (c) Brookhart, M.; Lincoln, D. M.; Bennett, M. A.; Pelling, S.
J. Am. Chem. Soc. 1990, 112, 2691. (d) Suzuki, H.; Kakigano, T.; Igarashi,
M.; Moro-oka, Y. Organometallics 1990, 9, 2192.
S0002-7863(98)00534-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/27/1998

6746 J. Am. Chem. Soc., Vol. 120, No. 27, 1998
EVans et al.
AgBPh4, eq 1. Bruno demonstrated that lanthanocene cations
routes to [(C5Me5)2Ln][BPh4] which did not depend on divalent
(C5Me5)2Ln precursors. We report here some new routes which
were developed with samarium and then applied to neodymium
and thulium. The reactivity of the neodymium analogue, [(C5-
Me5)2Nd][BPh4], was subsequently examined, and it was found
to provide the best synthesis to date for (C5Me5)3Nd.
(C5Me5)2Sm(THF)2 + AgBPh4 THF8
[(C5Me5)2Sm(THF)2][BPh4] + Ag (1)
of cerium and lanthanum could be prepared via abstraction of
iodide from trivalent [1,3-C5H3(SiMe3)2]2LnI(NCMe)2 with
AgBF4 or AgBPh4 to produce {[1,3-C5H3(SiMe3)2]2Ln(NCMe)2}-
{BF4} and {[1,3-C5H3(SiMe3)2]2Ln(NCMe)(DME)}[BPh4] (Ln
) Ce, La).8 Teuben prepared [(C5Me5)2CeL2][BPh4] [L ) THF,
tetrahydrothiophene (THT)] by protonoylsis of (C5Me5)2Ce[CH-
(SiMe3)2] with Et3NHBPh4 in THF or THT.9 Schaverien used
the same protonolysis methodology to prepare the monocyclo-
pentadienyl cation {(C5Me5)La[CH(SiMe3)2]}[BPh4] from (C5-
Me5)La[CH(SiMe3)2]2 and [PhNMe2H]BPh4.10 Very recently
the ytterbium complex [(Me3CC5H4)2Yb(THF)][BPh4]11 has
been prepared from the reduction of AgBPh4 with divalent (Me3-
CC5H4)2Yb(THF)2 in a synthesis analogous to that of [(C5Me5)2-
Sm(THF)2][BPh4]. Cationic lanthanide complexes based on
polypyrazolylborate ligands have also been described.12
Although many metallocene cations are more reactive due
to their increased electrophilicity, in the case of the solvated
complex [(C5Me5)2Sm(THF)2][BPh4] this led to tight binding
of the THF ligands which made the overall complex rather
unreactive.6 The solvated THF molecules were difficult to
displace, even in refluxing pyridine, and they blocked access
to the metal center, which resulted in limited reactivity.
However, the stability of [(C5Me5)2Sm(THF)2][BPh4] suggested
that the unsolvated [(C5Me5)2Ln]+ lanthanocene cations should
be highly reactive.
Initial attempts to make a solvent free cationic species from
unsolvated decamethylsamarocene, (C5Me5)2Sm,13 and AgBPh4
in toluene in a THF-free reaction analogous to eq 1 led to
insoluble materials. However, we now report that with the
proper procedures the soluble unsolvated cation complex, [(C5-
Me5)2Sm][BPh4] (1), can be isolated, and structurally character-
ized.
As anticipated, 1 is highly reactive and has proven to be a
good precursor to other trivalent samarium complexes. Among
its reactions is a new synthesis of (C5Me5)3Sm,14 a molecule
originally thought to be too sterically crowded to exist. The
synthetic utility of 1 indicated that this type of unsolvated
lanthanocene cation should be a valuable starting material for
other lanthanides. However, since the synthesis of 1 depended
on the divalent precursor, (C5Me5)2Sm, cations of lanthanides
which did not have a reactive divalent state could not be made
by this route. Accordingly, we sought to develop new synthetic
Experimental Section
The complexes described below are extremely air and moisture
sensitive. Therefore, both the syntheses and manipulations of these
compounds were conducted under nitrogen or argon with rigorous
exclusion of air and water by Schlenk, vacuum line, and glovebox
techniques. The argon glovebox used in these experiments was free
of coordinating solvents. The preparation of (C5Me5)2Sm,13 (C5Me5)2-
LnCl2K(THF)2,15,16 and (C5Me5)2Ln[CH(SiMe3)2]17 (Ln ) Sm, Nd, Tm)
and methods for drying solvents and taking physical measurements have
been described previously.18,19 LiCH(SiMe3)2 was prepared according
to literature procedure.20 Allylmagnesium chloride (2.0 M in THF,
Aldrich) was used as received. 1H and13C NMR analyses were carried
out on DRX 500 and 400 spectrometers. Infrared analyses were carried
out on KBr pellets with a Perkin-Elmer 1600 series FTIR. All C and
H elemental analyses were conducted on a Carlo Erba instrument. All
metal analyses were determined by complexometric titration.19
Synthesis of [(C5Me5)2Sm][BPh4] (1) from (C5Me5)2Sm. In an
argon-filled glovebox, a green solution of (C5Me5)2Sm (124 mg, 0.295
mmol) in ca. 10 mL of toluene was added to a slurry of AgBPh4 (426
mg, 0.298 mmol) in ca. 5 mL of toluene. The solution developed a
brown-red color in under 10 min. After the mixture was stirred for 12
h, a pale brown toluene solution was separated from dark solids by
centrifugation and discarded. The solids were extracted with ca. 20
mL of benzene to form a crimson red solution. Benzene was removed
by rotary evaporation and the solids were recrystallized from hot toluene
leaving red needles. The needles were dissolved in benzene and the
solvent was removed to give pure 1 as a rose-colored solid (137 mg,
63%). 1H NMR (C6D6) δ 8.4 (s, ∆ν1/2) 52 Hz, 8 H), -0.34 (s, C5Me5,
30H). 13C NMR (C6D6) δ 138, 131, 129, 124 (Ph) 120 (C5Me5), 21.9
(C5Me5). IR 3050 s, 2900 s, 1954 w, 1882 w, 1820 w, 1717 w, 1578
m, 1472 s, 1429 s, 1379 m, 1263 m, 1155 m, 1062 m, 1027 s, 959 m,
923 w, 872 m, 850 m, 805 w, 739 s, 707 s, 682 s, 610 m cm-1. Anal.
Calcd for SmC44H50B: Sm, 20.32; C, 71.41; H, 6.81. Found: Sm 20.0;
C, 70.5; H, 6.98. Addition of THF to 1 generated [(C5Me5)2Sm(THF)2]-
1
[BPh4] in quantitative yield, as observed by H NMR spectroscopy in
THF-d8.6 X-ray quality crystals of 1 were grown from a concentrated
benzene solution as red cubes.
Synthesis of 1 from (C5Me5)2Sm[CH(SiMe3)2] and Et3NHBPh4.
In an argon-filled glovebox, (C5Me5)2Sm[CH(SiMe3)2] (20 mg, 0.034
mmol), Et3NHBPh4 (15 mg, 0.036 mmol), ca. 15 mL of toluene, and
a stir bar were placed in a flask fitted with a high-vacuum greaseless
stopcock. The flask was attached to a Schlenk line via a water-cooled
condenser. The flask was placed under nitrogen and heated to reflux
and stirred for ca. 20 min. In that time the color of the solution turned
from orange to red. The solvent was removed leaving red solids. The
flask was returned to the argon glovebox, solids were extracted with
benzene, and solvent was removed to give 1 (23 mg, 90%). No
evidence for Et3N was observed in the 1H NMR spectrum on 1.
Attempts to increase the scale of the reaction to >100 mg resulted in
yields as low as 50%.
(5) (a) Marks, T. J.; Yang, X.; Stern, C. L. Organometallics 1991, 10,
840. (b) Marks, T. J.; Lin, Z.; Le Marechal, J.-F.; Sabat, M. J. Am. Chem.
Soc. 1987, 109, 4127.
(6) Evans, W. J.; Ulibarri, T, A.; Chamberlain, L. R.; Ziller, J. W.;
Alvarez, D., Jr. Organometallics 1990, 9, 2124.
(7) Evans, W. J.; Grate, J. W.; Choi, H. W.; Bloom, I.; Hunter, W. E.;
Atwood, J. L. J. Am. Chem. Soc. 1985, 107, 941.
(8) Hazin, P. N.; Bruno, J. W.; Schulte, G. K. Organometallics 1990, 9,
416.
Synthesis of (C5Me5)2Sm(η3-CH2CHCH2) (2) from ClMg-
(CH2CHCH2). In a nitrogen-filled glovebox, a 2.0 M THF solution
(9) Heeres, H. J.; Meetsma, J. H.; Teuben, J. H. J. Organomet. Chem.
1991, 414, 351.
(15) Evans, W. J.; Keyer, R. A.; Ziller, W. J. Organometallics 1993,
12, 2618.
(16) Evans, W. J.; Broomhall-Dillard, R. N. R.; Foster, S. E.; Ziller, J.
W. J. Coord. Chem. in press.
(17) Marks, T. J.; Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P.
N.; Schumann, H. J. Am. Chem. Soc. 1985, 107, 8091.
(18) Evans, W. J.; Grate, J. W.; Doedens, R. J. J. Am. Chem. Soc. 1985,
107, 1671.
(19) Evans, W. J.; Chamberlain, L. R.; Ulibarri, T. A.; Ziller, J. W. J.
Am. Chem. Soc. 1988, 110, 6423.
(10) Schaverien, C. J. Organometallics 1992, 11, 3476.
(11) Shen, Q.; Fugen, Y.; Sun, J. J. Organomet. Chem. 1997, 538, 241.
(12) Amoroso, A. J.; Jeffery, J. C.; Jones, P. L.; McCleverty, J. A.; Rees,
L.; Rheingold, A. L.; Sun, Y.; Takats, J.; Trofimenko, S.; Ward, M. D.;
Yap, G. P. A. J. Chem. Soc., Chem. Commun. 1995, 1881.
(13) (a) Evans, W. J.; Hughes, L. A.; Hanusa, T. P. Organometallics
1986, 5, 1285. (b) Evans, W. J.; Hughes, L.A.; Hanusa, T. P. J. Am. Chem.
Soc. 1984, 106, 4270.
(14) Evans, W. J.; Gonzales, S. L.; Ziller, J. W. J. Am. Chem. Soc. 1991,
113, 7423.
(20) Cowley, A. H.; Kemp, R. A. Synth. React. Inorg. Met.-Org. Chem.
1981, 11, 591.

UnsolVated Ln Metallocene Cations [(C5Me5)2Ln][BPh4]
J. Am. Chem. Soc., Vol. 120, No. 27, 1998 6747
of ClMg(CH2CHCH2) (1.5 mL, 3.0 mmol) was added to a stirring
suspension of (C5Me5)2SmCl2K(THF)2 (1.93 g, 2.86 mmol) in ca. 60
mL of toluene. The solution immediately turned deep red. After 30
min the solvent was removed by rotary evaporation leaving deep red
solids and a deep red viscous oil. A solution of dioxanes (ca. <1 vol
%) in hexanes was added, the mixture was stirred for ca. 30 min, and
the solution was filtered to remove white solids which were discarded.
Evaporation of solvent left red solids which were placed in a glass
tube fitted with a greaseless Teflon stopcock and heated at 60 °C at
10-7 Torr for ca. 12 h. The red/orange solids were transferred to an
argon-filled glovebox free of coordinating solvents and extracted with
hexanes. Removal of solvent by rotary evaporation left (C5Me5)2Sm-
(η3-CH2CHCH2) (2) (1.32 g, 87%), which was identified by 1H NMR
spectroscopy in benzene-d6.21
Synthesis of 1 from 2 and Et3NHBPh4. In an argon-filled
glovebox, (C5Me5)2Sm(η3-CH2CHCH2) (77 mg, 0.16 mmol) and Et3-
NHBPh4 (72 mg, 0.17 mmol) were combined with ca. 4-6 mL of
benzene and a stir bar. The light red solution was stirred at room
temperature. After 5 min, excess unreacted Et3NHBPh4 was removed
by centrifugation from the dark red solution. Rotary evaporation of
solvent left 1 (118 mg, 96%).
flask and the solution was stirred. Solids started to precipitate in less
than 10 min. The flask was stirred for 4 h, degassed, and transferred
to a glovebox where the solids were separated from the solvent by
centrifugation. The light blue/green solid was washed twice with small
amounts of hexanes then dried under vacuum leaving [(C5Me5)2Nd-
(µ-H)]2 (82 mg, 75%), which was identified by 1H NMR spectroscopy
in benzene-d6.17
Synthesis of [(C5Me5)2Nd][BPh4] (5). In an argon-filled glovebox,
3 (135 mg, 0.296 mmol) was reacted with Et3NHBPh4 (127 mg, 0.301
mmol) as described above for 1. A mint green benzene solution resulted
and 5 was isolated as a light green solid (211 mg, 97%) following the
procedure for 1. X-ray quality crystals of 5 were grown from a hot
toluene solution as green needles. 1H NMR (C6D6) δ 14.3 (s, ∆ν1/2
)
130 Hz). IR 3046 s, 2907 s, 1944 w, 1882 w, 1820 w, 1713 w, 1605
m, 1472 s, 1429 s, 1380 m, 1260 m, 1154 m, 1072 m, 1025 s, 949 w,
851 m, 805 w, 740 s, 706 s, 610 m cm-1
. Anal. Calcd for
NdC44H50B: Nd, 19.65; C, 72.01; H, 6.87. Found: Nd, 19.9; C, 71.5;
H, 6.56.
Synthesis of (C5Me5)2Nd[CH(SiMe3)2] from 5 and LiCH(SiMe3)2.
In an argon-filled glovebox, LiCH(SiMe3)2 (7 mg, 0.04 mmol) in ca.
3 mL of benzene was added to a blue-green solution of 5 (32 mg,
0.043 mmol) in ca. 5 mL of benzene. The solution turned blue in less
than 10 min and white solids were deposited. The solids were removed
by centrifugation and evaporation of solvent left (C5Me5)2Nd[CH-
(SiMe3)2] (24 mg, 96%) identified by 1H NMR spectroscopy in C6D6.17
Synthesis of [(C5Me5)2Tm][BPh4] (6). In an argon-filled glovebox,
4 (302 mg, 0.629 mmol) was reacted with Et3NHBPh4 (266 mg, 0.632
mmol) as described above for 1. An orange/yellow benzene solution
resulted. Evaporation of the solvent left an orange solid (467 mg).
Recrystallization from hot toluene left pale yellow needles and a pale
yellow solution. The crystals were removed by centrifugation and
washed with hexanes. The crystals were then dissolved in benzene
and evaporation of the solvent left [(C5Me5)2Tm][BPh4] (6) (298 mg,
63%) as an orange solid. IR 3049 s, 2898 s, 1945 w, 1871 w, 1825 w,
1713 w, 1578 m, 1473 s, 1429 s, 1384 m, 1265 m, 1155 m, 1071 m,
1027 s, 950 w, 913 w, 852 m, 848 m, 803 w, 738 s, 703 s, 677 s, 610
m cm.-1 Anal. Calcd for TmC44H50B: Tm, 22.3; C, 69.7; H, 6.64.
Found: Tm 21.4; C, 69.3; H, 6.42.
Synthesis of (C5Me5)3Nd (7). In an argon-filled glovebox, KC5-
Me5 (84 mg, 0.48 mmol) was reacted with 5 (323 mg, 0.44 mmol) as
described above for (C5Me5)3Sm, yielding (C5Me5)3Nd (7) (241 mg,
95%) as a mustard green solid. 1H NMR (C6D6) δ 8.88 (C5Me5). 13C
NMR (C6D6) δ 253 (C5Me5), δ -15.8 (C5Me5). IR 2904 vs, 1436 s,
1380 s, 1261 m, 1018 m br, 802 m, 597 m cm-1. Anal. Calcd for
NdC30H45: Nd, 26.23; C, 65.52; H, 8.25. Found: Nd, 26.0; C, 65.8;
H, 8.46. Dark orange X-ray quality crystals of 7 were grown from a
hot toluene solution.
Synthesis of (C5Me5)2Sm[CH(SiMe3)2] from 1 and LiCH(SiMe3)2.
In an argon-filled glovebox, LiCH(SiMe3)2 (7 mg, 0.04 mmol) in ca.
3 mL of benzene was added to a red solution of 1 (30 mg, 0.041 mmol)
in ca. 5 mL of benzene. The solution turned orange in less than 10
min and white solids precipitated. The solids were removed by
centrifugation, and evaporation of solvent left (C5Me5)2Sm[CH(SiMe3)2]
1
(22 mg, 94%) identified by H NMR spectroscopy.17
Synthesis of (C5Me5)2Sm from 1 and Na. In an argon-filled
glovebox, 1 (62 mg, 0.084 mmol) and ca. a 10 M excess of Na were
combined along with a stir bar in ca. 10 mL of benzene. The color of
the solution changed from red to green over a period of 12 h. White
solids (presumed to be NaBPh4), along with excess Na, were removed
by filtration, and the solvent was removed by rotary evaporation to
give (C5Me5)2Sm as a forest green solid (30 mg, 85% yield). The
1
identity of (C5Me5)2Sm was confirmed by H and 13C NMR spectros-
copy.13
Synthesis of (C5Me5)3Sm from 1 and KC5Me5. In an argon-filled
glovebox, KC5Me5 (29 mg, 0.16 mmol) and 1 (120 mg, 0.162 mmol)
were combined along with a stir bar in ca. 10 mL of benzene and the
reaction was stirred for ca. 12 h, during which time the color of the
solution turned from red to red/brown. White solids were removed
from the solution by centrifugation. Rotary evaporation of solvent left
1
(C5Me5)3Sm and a small amount of C5Me5H identified by H NMR
spectroscopy. Recrystallization from toluene left (C5Me5)3Sm (89 mg,
91%).14
Synthesis of (C5Me5)2Nd(η3-CH2CHCH2) (3) from ClMg-
(CH2CHCH2). As described above for 2, (C5Me5)2NdCl2K(THF)2
(1.97 g, 2.95 mmol) was reacted with a slight excess of 2 M ClMg-
(CH2CHCH2) in THF (1.5 mL, 3.0 mmol) and a lime green toluene
solution resulted. (C5Me5)2Nd(η3-CH2CHCH2) (1.13 g, 84%) was
isolated as described above for 2 and identified by 1H NMR spectros-
copy in toluene-d8.17
X-ray Data Collection, Structure Determination, and Refinement
for [(C5Me5)2Sm][BPh4] (1). A red crystal of approximate dimensions
0.20 × 0.23 × 0.24 mm was mounted on a glass fiber and transferred
to a Siemens P4 diffractometer. The determination of Laue symmetry,
crystal cell, unit cell parameters, and the crystal’s orientation matrix
was carried out according to standard procedures.22 Intensity data were
collected at 158 K by using a 2θ/ω scan technique with Mo KR
radiation. The raw data were processed with a local version of
CARESS,23 which employs a modified version of the Lehman-Larsen
algorithm to obtain intensities and standard deviations from the
measured 96-step peak profiles. Subsequent calculations were carried
out with the SHELXTL program.24 All 8635 data were corrected for
decay (5%), absorption, and Lorentz and polarization effects and were
placed on an approximately absolute scale. The diffraction symmetry
was 2/m with systematic absences 0k0 for k ) 2n + 1 and h0l for l )
2n + 1. The centrosymmetric monoclinic space group P21/c [C52h;
No. 14] is therefore uniquely defined.
Synthesis of (C5Me5)2Tm(η3-CH2CHCH2) (4) from ClMg-
(CH2CHCH2). As described above for 2, (C5Me5)2TmCl2K(THF) (618
mg, 0.995 mmol) was reacted with a slight excess of 2 M ClMg(CH2-
CHCH2) in THF (0.5 mL, 1.0 mmol) and a light yellow toluene solution
resulted. Solvent was removed leaving white solids which were handled
as described above for 2, leaving (C5Me5)2Tm(η3-CH2CHCH2) (4) (302
mg, 63%) as a bright yellow powder. Anal. Calcd for TmC23H35: Tm,
35.16; C, 57.5; H, 7.34. Found: Tm 35.0; C, 57.2; H, 7.14. IR 2906
vs, 2856 vs, 1538 s, 1435 m, 1377 s, 1246 m, 1094 m, 1020 m, 773 s,
679 s cm-1
.
Synthesis of [(C5Me5)2Nd(µ-H)]2 from 3 and H2. In an argon-
filled glovebox, 3 (117 mg, 0.257 mmol) in ca. 15 mL of hexanes and
a stir bar were placed in a flask fitted with a high-vacuum greaseless
stopcock. The flask was attached to a vacuum line and evacuated to
the solvent vapor pressure. Excess H2 at 1 atm was admitted to the
The structure was solved by direct methods and refined on F2 by
full-matrix least-squares techniques. The analytical scattering factors
(22) XSCANS Software Users Guide, Version 2.1, Siemens Industrial
Automation, Inc.; Madison, WI, 1994.
(23) Broach, R. W.; Argonne National Laboratory, Illinois, 1978.
(24) Sheldrick, G. M.; Siemens Analytical X-ray Instruments, Inc.;
Madison, WI, 1994.
(21) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1990,
112, 2314.

6748 J. Am. Chem. Soc., Vol. 120, No. 27, 1998
EVans et al.
for neutral atoms were used throughout the analysis.25 Hydrogen atoms
were located from a series of difference Fourier maps and refined (x,y,z
and Uiso). At convergence, wR2 ) 0.0551 and GOF ) 1.032 for 615
variables refined against all 8330 unique data (as a comparison for
refinement on F, R1 ) 0.0261 for those 6587 data with F > 4.0σ(F)).
X-ray Data Collection for [(C5Me5)2Nd][BPh4] (5). A pale green
crystal of approximate dimensions 0.20 × 0.18 × 0.06 mm was
mounted on a glass fiber and transferred to a Siemens P4 diffractometer.
The data on crystals of 5 were sufficient to provide atom connectivity,
but not high precision metrical information due to the poor quality of
the data collected.
X-ray Data Collection, Structure Determination, and Refinement
for (C5Me5)3Nd (7). An orange crystal of approximate dimensions
0.30 × 0.23 × 0.23 mm was mounted on a glass fiber and transferred
to a Siemens CCD platform diffractometer. The determination of Laue
symmetry, crystal class, and unit cell parameters was carried out
according to standard procedures.26 Intensity data were collected at
158 K with Mo KR radiation. The raw frame data were processed
with SAINT27 and SADABS28 to yield the reflection data file.
Subsequent calculations were carried out with the SHELXTL program.24
The diffraction symmetry was 6/m and the systematic absences are
consistent with the hexagonal space groups P63 or P63/m. It was later
determined that the centrosymmetric space group P63/m was correct.
The structure was solved by direct methods and refined on F2 by
full-matrix least-squares techniques. The analytical scattering factors
for neutral atoms were used throughout the analysis.25 Hydrogen atoms
were located from a difference Fourier map and refined (x,y,z and Uiso).
At convergence, wR2 ) 0.0339 and GOF ) 1.170 for 87 variables
refined against 1164 unique data (as a comparison for refinement on
F, R1 ) 0.0136 for those 1138 data with F > 4.0σ(F)).
Figure 1. Thermal ellipsoid plot of [(C5Me5)2Sm][BPh4] (1) with the
probability ellipsoids drawn at the 50% level. Hydrogen atoms have
been excluded for clarity.
Table 1. Experimental Data for the X-ray Diffraction Studies of
[(C5Me5)2Sm][BPh4] (1), [(C5Me5)2Nd][BPh4] (5), and (C5Me5)3Nd
(7)
1
5
7
formula
fw
temp (K)
crystal system
space group
a (Å)
C44H50BSm C51H58BNd C30H45Nd
740.00
158
826.02
158
549.90
158
monoclinic monoclinic hexagonal
P21/c
15.248(2)
14.2576(12) 28.144(7)
17.804(2) 13.277(3)
110.717(7) 98.54(2)
P21
11.200(3)
P63/m
10.0178(4)
10.0178(4)
15.5251(8)
120
b (Å)
c (Å)
Results
â (deg)
V (Å3)
Z
Synthesis of [(C5Me5)2Sm][BPh4] (1) from Divalent
(C5Me5)2Sm. In a reaction that parallels the formation of the
solvated cation, [(C5Me5)2Sm(THF)2][BPh4], from (C5Me5)2-
Sm(THF)2 and AgBPh4 in THF,6 eq 1, the THF-free samari-
um(II) complex (C5Me5)2Sm reacts with AgBPh4 in toluene over
a 12 h period according to eq 2. Although the initially formed
3620.4(6)
4
1.358
4139(2)
4
1.326
1349.30(10)
2
1.938
D
calcd (Mg/m3)
diffractometer
Siemens P4 Siemens P4 Siemens CCD
µ(Mo KR) (mm-1
)
1.651
1.288
0.1193
1.938
0.0339
refinementb wR2 (all data) 0.0551
a Radiation: Mo KR (λ ) 0.710730 Å). Monochromator: highly
oriented graphite. b R1 ) [∑[w(Fo - Fc2)2]/∑(w(Fo )2)]1/2
.
2
2
(C5Me5)2Sm + AgBPh4 toluene8
black toluene insoluble solids benzene8
[(C5Me5)2Sm][BPh4] + black solids (2)
Characterization of [(C5Me5)2Sm][BPh4]. 1 was character-
ized by IR and NMR spectroscopy, elemental analysis, and
addition of THF to produce the previously reported solvated
complex [(C5Me5)2Sm(THF)2][BPh4]6 in quantitative yield. The
structure of 1 was determined by crystal structure analysis and
found to have the solid state arrangement shown in Figure 1.
black product is not noticeably soluble in toluene at room
temperature, extraction with benzene provided a crimson red
solution (ca. 18 mg/mL) of [(C5Me5)2Sm][BPh4] (1). Yields
of 60-65% were obtained in toluene at room temperature.
Reactions conducted in benzene at room temperature and at
reflux and reactions in toluene at reflux and -78 °C did not
give higher yields of 1.
It is crucial that the starting materials be meticulously free
of any coordinating solvents, since 1 readily adds Lewis bases.
The purity of AgBPh4, which requires an aqueous synthesis, is
especially important. Since 1 is soluble in hot toluene and forms
red needles upon cooling, hot toluene extraction provides a
convenient purification method. Unfortunately, these crystals
were of insufficient quality for X-ray diffraction analysis.
Recrystallization from benzene, on the other hand, provided 1
as red cubes suitable for X-ray diffraction.
1
The H NMR spectrum of paramagnetic 1 in C6D6 was not
structurally definitive. The spectrum contained a singlet at δ
-0.34, assigned as the methyl protons on the C5Me5 rings, and
a broad signal at δ 8.4 that integrated for 7 to 8 protons
compared to 30 for the signal at δ -0.34. Low-temperature
NMR studies were precluded by the low solubility of 1 in
toluene-d8 and the freezing point of C6D6. Both high-temper-
ature and spin-lattice (T1)29 experiments produced spectra
identical to the room-temperature spectrum. The shifts of the
C5Me5 carbon atoms in the 13C NMR spectrum of 1 were
consistent with the presence of Sm(III),30 while the four
remaining signals could be assigned to phenyl carbons.
In the X-ray crystal structure of 1, the [(C5Me5)2Sm]+ cation
is oriented toward two of the phenyl rings of the [BPh4]-
counteranion. Selected bond angles and distances are shown
in Table 2. The 2.70(2) Å average Sm-C(C5Me5) distance is
indistinguishable from the 2.69(2) Å average in [(C5Me5)2Sm-
(25) International Tables of X-ray Crystallography; Kluwer Academic
Publishers: Dordrecht, 1992; Vol. C.
(26) SMART Software Users Guide, Version 4.21; Siemens Industrial
Automation, Inc.; Madison, WI, 1996.
(27) SAINT Software Users Guide, Version 4.05; Siemens Industrial
Automation, Inc.; Madison, WI, 1996.
(28) Sheldrick, G. M. SADABS; Siemens Analytical X-ray Instruments,
Inc.; Madison, WI 1996.
(29) Marks, T. J.; Fischer, R. D. Organometallics of the f-Elements;
Proceedings of the NATO Advanced Study Institute; D. Reidel: Boston,
MA, 1978; Chapter 11.
(30) Evans, W. J.; Ulibarri, T. A. J. Am. Chem. Soc. 1987, 109, 4292.

UnsolVated Ln Metallocene Cations [(C5Me5)2Ln][BPh4]
J. Am. Chem. Soc., Vol. 120, No. 27, 1998 6749
actinides.35 Comparison of these distances can also be made
with the structure of [(C5Me5)2ThMe][B(C6F5)4],5a which has
2.675(5) and 2.757(4) Å Th‚‚‚F, 2.754(3) Å Th-C(C5Me5), and
2.399(8) Å Th-C(Me) distances (7-coordinate Th4+ is 0.037
Å larger than 6-coordinate Sm3+ based on Shannon radii36).
Table 2. Selected Bond Distances (Å) and Angles (deg) for
[(C5Me5)2Sm][BPh4] (1) and (C5Me5)3Nd (7)
1
7
Sm(1)-C(1)
Sm(1)-C(2)
Sm(1)-C(3)
Sm(1)-C(4)
Sm(1)-C(5)
Sm(1)-C(11)
Sm(1)-C(12)
Sm(1)-C(13)
Sm(1)-C(14)
Sm(1)-C(15)
2.704(3) Nd(1)-C(1)
2.740(3) Nd(1)-C(2)
2.721(3) Nd(1)-C(3)
2.8146(13)
2.8421(14)
2.927(2)
The orientations of the planes of the two phenyl rings closest
to the samarium metal center were evaluated by measuring the
dihedral angles between the Sm, C(22), C(23) plane and the
C(22), C(23), phenyl ring centroid plane (124°) and the Sm,
C(28), C(29) plane and the C(28), C(29), phenyl ring centroid
plane (127°). These angles show that the rings interact at an
angle with the metal rather than with the edge or face of the
ring.
2.696(3) Nd(1)-C(C5Me5)d 2.86(6)
2.670(3) Nd(1)-Cente
2.722(3)
2.582
2.675(2) Cnt-Nd(1)-Cnt
120
2.686(3) C(2)-C(1)-C(4) 122.7(2)
2.703(3) C(1)-C(2)-C(5) 123.7(2)
2.735(3) C(2)-C(3)-C(6) 124.64(9)
Sm(1)-C(C5Me5)a
Sm(1)-C(22)
Sm(1)-C(28)
Sm(1)-C(23)
Sm(1)-C(29)
2.70(2)
2.825(3)
2.917(3)
3.059(3)
3.175(3)
The interactions of the phenyl rings with the samarium cation
do not change the tetrahedral nature of the phenyl rings around
boron: the six C-B-C angles are in the narrow range of 107.5-
(2)-110.9(2)o. The C(phenyl)-C(phenyl) bond distances within
the two closest phenyl rings have a narrow range, 1.381(4)-
1.404(2) Å, similar to that in the other two phenyl rings, 1.375-
(4)-1.405(4) Å, and no localization is observed.
Cnt(1)b-Sm(1)-Cnt(2)c 134.4(2)
a Sm(1)-C(C5Me5) is the average bond distance between Sm(1) and
C(1) to C(5) and C(11) to C(15). b Cnt(1) is the centroid of the C(1)-
C(5) ring. c Cnt(2) is the centroid of the C(11)-C(15) ring. d Nd(1)-
C(C5Me5) is the average bond distance between Nd(1) and C(1) to C(3).
e Cnt is the centroid of the C(1), C(2), C(3), C(1C), C(2C) ring.
Reactivity of [(C5Me5)2Sm][BPh4]. The reactivity of 1 was
initially compared to that of the solvated analogue [(C5Me5)2-
Sm(THF)2][BPh4].6 [(C5Me5)2Sm(THF)2][BPh4] can be reduced
with sodium to form the divalent solvated metallocene (C5Me5)2-
Sm(THF)2 as shown in eq 3. 1 is similarly reduced, except
that the unsolvated divalent metallocene, (C5Me5)2Sm,13 is
isolated instead, eq 4.
(THF)2][BPh4].6 Both of these averages are on the low end of
the range of distances found in trivalent (C5Me5)2Sm-containing
complexes,31 as might be expected for a complex with a net
positive charge and enhanced electrophilicity. The 134.4° (ring
centroid)-Sm-(ring centroid) angle is normal and matches the
134.2° angle in [(C5Me5)2Sm(THF)2][BPh4].6 1 and [(C5Me5)2-
Sm(THF)2][BPh4] differ in that the rings are staggered in 1 and
eclipsed in the solvated analogue. This latter orientation may
be due to the steric crowding caused by the closely located and
tightly held THF ligands (Sm-O(THF) ) 2.46(1) Å).6
[(C5Me5)2Sm(THF)2][BPh4] + Na THF8
(C5Me5)2Sm(THF)2 + NaBPh4 (3)
[(C5Me5)2Sm][BPh4] + Na benzene8
(C5Me5)2Sm + NaBPh4 (4)
The closest approach of the [BPh4]- anion to samarium is
through the o-carbon atoms C(22) and C(28) at distances of
2.825(3) and 2.917(3) Å, respectively. m-Carbon atoms C(23)
and C(29) are located 3.059(3) and 3.175(3) Å from Sm,
respectively. In comparison, the Sm-C(phenyl) single bond
distance in (C5Me5)2Sm(C6H5)(THF)32 is 2.511(8) Å and the
Sm-C(methyl) single bond distance in (C5Me5)2Sm(CH3)-
(THF)19 is 2.484(14) Å. Since the Sm-C(o-phenyl) distances
in 1 are longer than Sm-C single bonds and the Sm-C(ring)
distances, they represent weaker interactions than these normal
connections. Similar “long range” interactions have been found
in other organosamarium complexes such as [(C5Me5)2Sm(Ct
CCMe3)]2, [(C5Me5)2Sm][µ-η2:η2-Ph(CH2)2CdCdCdC(CH2)2-
Ph],15 and (C5Me4Et)3Sm,33 and a wealth of agostic interactions
have been reported for electrophilic organolanthanides17,34 and
1 and [(C5Me5)2Sm(THF)2][BPh4] also have similar reactivity
with LiCH(SiMe3)2: both produce (C5Me5)2Sm[CH(SiMe3)2]
in quantitative yields,37 eq 5. In this case, since the product is
[(C5Me5)2Sm(THF)2][BPh4] + LiCH(SiMe3)2 THF8
[(C5Me5)2Sm][BPh4] + LiCH(SiMe3)2 benzene8
}
(C5Me5)2Sm{CH(SiMe3)2} + LiBPh4 (5)
THF free, the same result is obtained with both cations. This
type of reaction between 1 and alkali metal anion, MZ (M )
alkali metal, Z ) anion), is likely to be a convenient route to
many types of (C5Me5)2SmZ complexes.
(31) Evans, W. J.; Foster, S. E. J. Organomet. Chem. 1992, 433, 79.
(32) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. Organome-
tallics 1985, 4, 112.
(33) Evans, W. J.; Forrestal, K. J.; Leman, J. T.; Ziller, J. W.
Organometallics 1996, 15, 527.
A reaction that would necessarily be very different for 1 and
[(C5Me5)2Sm(THF)2][BPh4] is the reaction with KC5Me5. In
the solvated case, this reaction does not give the (C5Me5)2SmZ
product, which would be (C5Me5)3Sm, but instead produces the
THF ring-opened product, (C5Me5)2Sm[O(CH2)4C5Me5], eq 6.6
Since 1 is THF free, this reaction is not possible and the reaction
(34) (a) Teuben, J. H.; Duchateau, R.; van Wee, C. T. Organometallics
1996, 15, 2291. (b) Takats, J.; Zhang, X.; McDonald, R. New J. Chem.
1995, 19, 573. (c) Marks, T. J.; Rheingold, A. L.; Giardello, M. A.;
Conticello, V. P.; Brard, L.; Sabat, M.; Stern, C. L. J. Am. Chem. Soc.
1994, 116, 10212. (d) Takats, J.; Hasinoff, L.; Zhang, X. W. J. Am. Chem.
Soc. 1994, 116, 8833. (e) Teuben, J. H.; Duchateau, R.; van Wee, C. T.;
Meetsma, A. J. Am. Chem. Soc. 1993, 115, 4931. (f) Teuben, J. H.; Booij,
M.; Deelman, B.-J.; Duchateau, R.; Postma, D. S.; Meetsma, A. Organo-
metallics 1993, 12, 3531. (g) Teuben, J. H.; Booij, M.; Meetsma, A.
Organometallics 1991, 10, 3246. (h) Teuben, J. H.; Heeres, H. J.; Meetsma,
A. Organometallics 1989, 8, 2637. (i) Teuben, J. H.; Martin, B.; Kiers, N.
H.; Meetsma, A.; Smeets, W. J. J.; Spek, A. L. Organometallics 1989, 8,
2454. (j) Teuben, J. H.; Heeres, H. J.; Renkema, J.; Booij, M.; Meetsma,
A. Organometallics 1988, 7, 2495. (k) Teuben, J. H.; Den Haan, K. H.; De
Boer, J. L. J. Organomet. Chem. 1987, 327, 31.
(35) (a) Takats, J.; Sun, Y.; Eberspacher, T.; Day, V. Inorg. Chim. Acta
1995, 229, 315. (b) Takats, J.; Sun, Y.; McDonald, R.; Day, V. W.;
Eberspacher, T. A. Inorg. Chem. 1994, 33, 4433.
(36) Shannon, R. D. Acta Crystallogr. 1976, A32, 751.
(37) (C5Me5)2Sm[CH(SiMe3)2] has previously been prepared by other
routes, but in low yields (45%).17

6750 J. Am. Chem. Soc., Vol. 120, No. 27, 1998
EVans et al.
toluene
THF
[(C5Me5)2Sm(THF)2][BPh4] + KC5Me5
(C5Me5)2Sm[CH(SiMe3)2] + Et3NHBPh4
8
reflux
Et3N + H2C(SiMe3)2 + [(C5Me5)2Sm][BPh4] (11)
(C5Me5)2Sm(THF)O
(6)
In these cases, dark solids, which proved to be insoluble even
in THF, were formed. The typical decomposition products C5-
Me5H or [(C5Me5)2Sm]2(µ-O)41 were not observed in the H
1
with KC5Me5 provides a new route to (C5Me5)3Sm. This reaction
occurs in ca. 90% yield as shown in eq 7.
NMR spectra of these reactions and could not explain the low
yields. One important observation from these reactions was
that the Et3N byproduct was easily removed with solvent during
evaporation leaving amine free 1.
An attractive alternative to (C5Me5)2Sm[CH(SiMe3)2] was the
allyl complex (C5Me5)2Sm(η3-CH2CHCH2) (2), since (a) it is
also free of coordinating solvents, (b) the byproduct of the
reaction, propylene, could be easily removed from the reaction
in situ, and (c) it can be easily purified by extraction with
hexanes. As shown in eq 12, 1 can be prepared in quantitative
[(C5Me5)2Sm][BPh4] + KC5Me5 benzene8
(C5Me5)3Sm + KBPh4 (7)
The formation of (C5Me5)3Sm from 1 is an important
synthetic breakthrough in the chemistry of the sterically crowded
(C5Me5)3Ln complexes. The first two syntheses reported for
(C5Me5)3Sm, eqs 819 and 9,33 required divalent starting materials,
(C5Me5)2Sm(η3-CH2CHCH2) + Et3NHBPh4 benzene8
Et3N + CH3CHdCH2 + [(C5Me5)2Sm][BPh4] (12)
2(C5Me5)2Sm + C8H8 toluene8 (C5Me5)3Sm +
(C5Me5)Sm(C8H8) (8)
yield from 2 and Et3NHBPh4 in benzene at ambient tempera-
tures. The reaction can be carried out with excess Et3NHBPh4
with no decomposition due to protonation of C5Me5 ligands.
The synthesis of 1 from 2 is much more efficient in time, yield,
and cost when compared to the (C5Me5)2Sm[CH(SiMe3)2]-based
synthesis and the reaction can be conducted successfully on a
large scale.
2(C5Me5)2Sm(OEt)2 + (C5Me5)2Pb toluene8
2(C5Me5)3Sm + Pb (9)
(C5Me5)2Sm and (C5Me5)2Sm(OEt2)2.38 This precluded exten-
sion of these synthetic methods to the other lanthanides, since
none have a divalent metallocene as reactive as (C5Me5)2Sm.
(C5Me5)3Sm can also be made from the trivalent precursor [(C5-
Me5)2SmH]2,39 eq 10,40 which is readily made from the divalent
New Syntheses of (C5Me5)2Ln(η3-CH2CHCH2) from Triva-
lent Lanthanide Precursors. Although 2 proved to be a good
precursor to 1, the best route to 2 is from divalent (C5Me5)2-
Sm, eq 13.42 (C5Me5)2Ln(η3-CH2CHCH2) complexes, including
2(C5Me5)2Sm + 3CH3CHdCH2 f CH3CH2CH3 +
2(C5Me5)2Sm(η3-CH2CHCH2) (13)
toluene
1/2 [(C5Me5)2SmH]2 +
(C5Me5)3Sm
(10)
the samarium complex, can be made from trivalent precursors,
but this requires [(C5Me5)2LnH]n,17,42 eq 14. Since the trivalent
(C5Me5)2Sm(THF)2, but analogous syntheses of (C5Me5)3Ln
complexes require the preparation of [(C5Me5)2LnH]2 from
trivalent precursors, which is neither an easy nor a high-yield
process.17 Since the unsolvated cations, [(C5Me5)2Ln][BPh4],
would be much better precursors to (C5Me5)3Ln complexes than
any other currently known precursors, syntheses of [(C5Me5)2-
Ln][BPh4] complexes which did not depend on divalent (C5-
Me5)2Ln were sought. Initial studies used samarium since there
is more NMR data available on (C5Me5)2Ln-containing com-
plexes of samarium than for any other lanthanide.
Synthesis of [(C5Me5)2Sm][BPh4] from Trivalent (C5Me5)2-
SmR Precursors. The formation of 1 by protonolysis of (C5-
Me5)2Sm[CH(SiMe3)2] was first investigated because this
precursor has been fully characterized17 and can be readily
isolated without coordinated solvent from trivalent starting
materials. It was expected that CH(SiMe3)2 could be selectively
protonated over C5Me5 based on the preparation of {(C5-
Me5)La[CH(SiMe3)2]}[BPh4] from (C5Me5)La[CH(SiMe3)2]2.10
Indeed, 1 can be prepared from the reaction of (C5Me5)2Sm-
[CH(SiMe3)2] with Et3NHBPh4 in refluxing toluene in yields
from 50 to 90% depending on scale of reaction, eq 11.
Unfortunately, the yield is lower for the larger scale reactions.
1/2[(C5Me5)2SmH]2 + 2CH3CHdCH2 f CH3CH2CH3 +
(C5Me5)2Sm(η3-CH2CHCH2) (14)
route to prepare these hydrides comes from (C5Me5)2Ln[CH-
(SiMe3)2], allyl complexes prepared from [(C5Me5)2LnH]n offer
no synthetic advantage. Accordingly, a simpler and more direct
route to THF-free (C5Me5)2Ln(η3-CH2CHCH2) complexes had
to be developed. Again, this was investigated first with
samarium.
Since it had been reported that the simple cyclopentadienyl
complex (C5H5)2Sm(CH2CHCH2)(THF)x could be prepared from
a reaction of (C5H5)2SmCl with allylmagnesium bromide in
THF,43 a similar route was examined with allylmagnesium
chloride and (C5Me5)2SmCl2K(THF)2, which is directly obtain-
able from SmCl3 and 2 equiv of KC5Me5.15 A toluene
suspension of (C5Me5)2SmCl2K(THF)2 reacts instantly with a
solution of ClMgCH2CHCH2 in THF. Unfortunately, initial
attempts to extract the products in hexanes failed to produce
any soluble materials. However, the addition of ca. 1 vol % of
dioxane in hexanes produced a dark red solution along with
(38) Berg, D. J.; Burns, C. J.; Andersen, R. A.; Zalkin, A. Organome-
tallics 1989, 8, 1865.
(39) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem.
Soc. 1983, 105, 1401.
(40) Evans, W. J.; Forrestal, K. J.; Ziller, J. W. Angew. Chem., Int. Ed.
Engl. 1997, 36, 774.
(41) Evans, W. J.; Grate, J. W.; Bloom, I.; Hunter, W. E.; Atwood, L.
J. J. Am. Chem. Soc. 1985, 107, 405.
(42) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1990,
112, 2314.
(43) Tsutsui, M.; Ely, N. J. Am. Chem. Soc. 1975, 97, 3551.

UnsolVated Ln Metallocene Cations [(C5Me5)2Ln][BPh4]
J. Am. Chem. Soc., Vol. 120, No. 27, 1998 6751
Scheme 1
white solids, presumed to be dioxane solvates of MgCl2 as well
as KCl, which were removable by filtration. Removal of the
dioxane/hexanes solvent at room temperature left the solvated
complex (C5Me5)2Sm(CH2CHCH2)(THF)n,21 along with traces
of dioxane. Heating the solids between 55 and 70 °C at 10-7
Torr over a 4-20 h period (depending on the amount of
material) followed by extraction in hexanes left pure 2 in yields
over 80% with no trace of dioxane or THF, Scheme 1. The
neodymium and thulium analogues, (C5Me5)2Ln(η3-CH2-
CHCH2) (Ln ) Nd (3), Tm (4)), were prepared analogously
and both 2 and 3 have been prepared on a gram scale with this
procedure.
Figure 2. Thermal ellipsoid plot of (C5Me5)3Nd (7) with the probability
ellipsoids drawn at the 50% level. Hydrogen atoms have been excluded
for clarity.
yields over 90%, eq 17. 7 was characterized by NMR and IR
+ LiCH(SiMe3)2 benzene8
[(C Me ) Nd][BPh ]
5
5 2
4
5
(C5Me5)2Nd{CH(SiMe3)2} + LiBPh4 (16)
+ KC5Me5 benzene8
4
[(C Me ) Nd][BPh ]
5
5 2
5
Synthesis of [(C5Me5)2Nd][BPh4] and [(C5Me5)2Tm][BPh4]
from (C5Me5)2Ln(η3-CH2CHCH2). (C5Me5)2Nd(η3-CH2-
CHCH2) (3) reacts with Et3NHBPh4 in benzene to produce [(C5-
Me5)2Nd][BPh4] (5) in quantitative yield, eq 15. As in the
(C Me ) Nd
+ KBPh4 (17)
5
5 3
7
spectroscopy as well as by a crystal structure determination to
unequivocally confirm that it is the third fully characterized
example of a (C5Me5)3M complex, Figure 2.
(C5Me5)2Ln(η3-CH2CHCH2) + Et3NHBPh4 benzene8
Et3N + CH3CHdCH2 + [(C5Me5)2Ln][BPh4] (15)
Complex 7 was found to be isostructural and isomorphous
with (C5Me5)3Sm.14 The differences in the three individual
Ln-C bond distances in the two complexes are consistent with
the 0.031 Å difference in ionic radii of the two metals:36 the
differences are 0.017(5) Å for Ln-C(1), 0.025(4) Å for Ln-
C(2), and 0.033(4) Å for Ln-C(3).
Ln ) Nd (5), Tm (6)
synthesis of 1, excess Et3NHBPh4 could be used with no
observable decomposition of 5. Likewise, the reaction of 4 with
Et3NHBPh4 in benzene produces [(C5Me5)2Tm][BPh4] (6).
However, a more modest yield, 63%, was observed for this
smaller metal.
The 1H NMR spectrum of 7 contains a singlet at δ 8.88 while
the 13C NMR spectrum contains singlets at δ 253 and -15.8
ppm for the ring and methyl carbons, respectively. 7 was first
prepared from a reaction of [(C5Me5)2NdH]2 with tetramethyl-
fulvalene, but was not obtained in sufficient yield for full
characterization.45
Characterization of [(C5Me5)2Nd][BPh4] (5) and [(C5Me5)2-
Tm][BPh4] (6). Both 5 and 6 were characterized with use of
analytical methods as well as IR spectroscopy. 5 was further
characterized by NMR spectroscopy and a crystal structure
determination. The IR spectra of 5 and 6 were nearly identical
to that of 1, while the NMR spectra of 5 were less informative
due to the larger magnetic moment of Nd(III) (3.5-3.6 µB).44
Unlike 1 and 5, characterization of 6 by NMR spectroscopy
was precluded by the large magnetic moment of trivalent
thulium (7.1-7.4 µB).44 The similar solubility properties of 5
and 6 compared to 1 again allowed for purification by
recrystallization from hot toluene. In contrast to 1, crystals of
5 or 6 suitable for X-ray diffraction were not obtained from
benzene. However, needles of 5 produced from recrystallization
in hot toluene were examined by X-ray diffraction and found
to be isostructural with 1.
Attempts to reduce 7 with either Na or K to form a divalent
complex in analogy to eq 4 were conducted due to the reported
preparation of a divalent neodymium complex, [K(THF)n][(C5-
Me5)2NdCl2].46 No new Nd(II) products were isolated and
mostly starting material was recovered.
Reaction of [(C5Me5)2Tm][BPh4] with Potassium. At-
tempts to form (C5Me5)2Tm from 6 by reduction with potassium
in analogy to eq 4 produced dark toluene insoluble products
which formed green/blue solutions in dimethoxyethane. Al-
though these dark colors are similar to that of the crystallo-
graphically characterized Tm(II) complex TmI2(DME)3,47 crys-
tallographic data on a new Tm(II) complex were not obtained.
Efficient Synthesis of [(C5Me5)2LnH]n from (C5Me5)2Ln-
(η3-CH2CHCH2). It was previously shown that the allyl
Reactivity of [(C5Me5)2Nd][BPh4]. Complex 5, like 1, reacts
with LiCH(SiMe3)2 in benzene to produce the known complex
(C5Me5)2Nd[CH(SiMe3)2]17 in quantitative yield, eq 16. 5 reacts
slowly with KC5Me5 in benzene to produce (C5Me5)3Nd, 7, in
(45) Forrestal, K. J. Ph.D. Dissertation, University of California, Irvine,
1997.
(46) Wedler, M.; Recknagel, A.; Edelmann, F. T. J. Organomet. Chem.
1990, 395, C26.
(47) Bochkarev, M. N.; Fedushkin, I. L.; Fagin, A. A.; Petrovskaya, T.
V.; Ziller, J. W.; Broomhall-Dillard, R. N. R.; Evans, W. J. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 113.
(44) Evans, W. J.; Hozbor, M. A. J. Organomet. Chem. 1987, 326, 299.

6752 J. Am. Chem. Soc., Vol. 120, No. 27, 1998
EVans et al.
complex (C5Me5)2Sm(η3-CH2CHCH2) could be hydrogenolyzed
to [(C5Me5)2SmH]2, eq 18.42 Likewise, (C5Me5)2Nd(η3-CH2-
on [(C5Me5)2LnH]n and its precursor (C5Me5)2Ln[CH(SiMe3)2].
Since these unsolvated allyl species can now be conveniently
made directly from LnCl3, KC5Me5, and ClMg(CH2CHCH2) in
high yield for samarium, neodymium, and thulium, it is likely
that these (C5Me5)2Ln(η3-CH2CHCH2) complexes will be more
heavily utilized as pseudo-alkyl starting materials in the future
not only to make cationic metallocenes, but also to make other
lanthanide metallocene derivatives. The allyl complexes are
cheaper and easier to prepare than the commonly used (C5Me5)2-
Ln[CH(SiMe3)2]17 and have been shown to readily form (C5-
Me5)2Sm(η1-CH2CHdCH2) intermediates.21,48,49 As shown for
neodymium, the improved synthesis now makes these allyl
complexes the best precursors to the [(C5Me5)2LnH]n hydrides
which, in the past, were the precursors to the allyl complexes
for Ln * Sm.
The development of the (C5Me5)2NdCl2K(THF)2 to (C5Me5)2-
Nd(η3-CH2CHCH2) to [(C5Me5)2Nd][BPh4] route for neody-
mium, which made it possible to prepare (C5Me5)3Nd for
definitive characterization, shows the utility of these results for
the synthesis of new complexes as well. Other (C5Me5)3Ln
complexes should be accessible by this route, and this should
extend the unusual chemistry of (C5Me5)3Sm14,50 to the other
lanthanide metals.
hexanes
(C5Me5)2Ln
+ H2
[(C5Me5)2LnH]n + CH2 CHCH3
(18)
17
CHCH2) reacts with H2 in hexanes to make [(C5Me5)2NdH]2
in 75% yield, eq 18. In the past, (C5Me5)2Nd(η3-CH2CHCH2)
was made from the hydride, which was made from (C5Me5)2-
Nd[CH(SiMe3)2], eq 19.17 Hence, the improved synthesis of
these allyl complexes also provides an improved synthesis of
the hydrides.
pentane
[(C5Me5)2NdH]2
2(C5Me5)2Nd
2(C5Me5)2Nd[CH(SiMe3)2] + 2H2
–2CH2(SiMe3)2
4CH2
CHCH3
(19)
–2CH3CH2CH3
Discussion
As anticipated, the unsolvated cation, [(C5Me5)2Sm]+, is
highly electrophilic. Fortunately, the electrophilicity of the
metal center can be satisfied by a single [BPh4]- unit, which
interacts strongly enough to generate a soluble zwitterion [(C5-
Me5)2Sm][BPh4] (1) rather than an insoluble, intermolecularly
connected oligomer involving bridging C5Me5 or BPh4 groups.
It is also fortunate that the [BPh4]- moiety can be easily
displaced by other reagents. Facile loss of [BPh4]- effectively
provides a soluble form of the highly electrophilic [(C5Me5)2-
Sm]+ cation. In this regard, complex 1 should become a
generally useful starting material in organosamarium chemistry.
This was best demonstrated by the fact that it can provide a
route to sterically crowded (C5Me5)3Sm, a molecule that in the
past has been difficult to prepare.
Although the synthesis of (C5Me5)3Sm from 1 is now the
fourth synthesis of (C5Me5)3Sm, it is the one that is best for
generalization to the other lanthanides. This is typical of the
development of organolanthanide metallocene chemistry. The
high reactivities of divalent reagents (C5Me5)2Sm(THF)2 and
(C5Me5)2Sm often provide the first examples of new types of
lanthanide metallocene derivatives and hence organosamarium
chemistry develops faster than that of the other elements. Once
a particular type of new complex is discovered, one can devise
routes to the analogous complexes of the other metals using
trivalent chemistry.
Conclusion
The unsolvated decamethylsamarocenium cation complex
[(C5Me5)2Sm][BPh4] was synthesized, structurally characterized,
and used to develop a general convenient synthetic route to
unsolvated [(C5Me5)2Ln][BPh4] cations of other lanthanides.
These cations should have wide applicability as precursors in
organolanthanide chemistry, since the tetraphenylborate anion
can be displaced by a variety of ligands. The [(C5Me5)2Ln]-
[BPh4] complexes currently are the best starting material for
the syntheses of (C5Me5)3Ln complexes via Ln(III) routes. In
connection with the development of the synthesis of these
lanthanocene cations, a convenient synthesis was also found
for the allyl complexes (C5Me5)2Ln(η3-CH2CHCH2), which
should make these complexes easily obtainable metallocene
pseudo-alkyl alternatives to the commonly used alkyls (C5Me5)2-
Ln[CH(SiMe3)2].
Acknowledgment. We thank the National Science Founda-
tion for support of this research.
Supporting Information Available: Atomic coordinates,
thermal parameters, and complete bond distances and angles
for compounds 1, 5, and 7 (37 pages, print/PDF). See any
current masthead page for ordering information and Web access
instructions.
To use [(C5Me5)2Ln][BPh4] complexes as precursors to (C5-
Me5)3Ln and other compounds where Ln * Sm, it was necessary
to develop a synthesis that did not rely on Ln(II). The general
synthesis of [(C5Me5)2Ln][BPh4] from trivalent (C5Me5)2Ln-
(η3-CH2CHCH2) provides this route and emphasizes the im-
portance of these allyl complexes as precursors. To use the
(C5Me5)2Ln(η3-CH2CHCH2) complexes as precursors in cases
in which Ln * Sm, a synthesis of the allyl compounds that did
not depend on Ln(II) was needed. In addition, it was desirable
to have a synthesis of the allyl complexes that did not depend
JA980534O
(48) Evans, W. J.; DeCoster, D. M.; Greaves, J. Macromolecules, 1995,
34, 5927.
(49) Evans, W. J.; Seibel, C. A.; Ziller, J. W. Organometallics 1998,
17, 2103.
(50) Evans, W. J.; Forrestal, K. J.; Ziller, J. W. Fifth Chemical Congress
of North America, November 11-15, 1997, Cancu`n, Mexico.