Actinide hydride complexes as multielectron reductants: Analogous reduction chemistry from [(C5Me5)2UH]2, [(C5Me5)2UH2]2, and [(C5Me5)2ThH2]2

Article abstract of DOI:10.1021/om7003139

Methods to separate the components of the equilibrium mixture of [(C ₅Me₅)₂UH]₂ and [(C ₅Me₅)₂-UH]₂ have been developed that allow their reductive chemistry to be studied. These actinide hydrides can act as four-, six-, and eight-electron reductants depending on the substrate with H: as the byproduct of a H^- → e^- + 1/2 H₂ redox couple. This hydride reduction chemistry allows complexes of redox-inactive Th⁴⁺ such as [(C₅Me₅) ₂THH₂]₂ to be four- and six-electron reductants. [(C₅Me₅)₂UH]₂ and [(C₅-Me₅)₂UH₂]₂ cleanly reduce 2 equiv of PhEEPh (E = S, Se) to form 2 equiv of (C₅Me ₅)₂U(SPh)₂ and (C₅Me ₅)₂U(SePh)₂ in an overall four-electron reduction in each case. [(C₅Me₅)₂UH] ₂ and [(C₅Me₅)₂UH₂] ₂ also effect a six-electron reduction of 3 equiv of 1,3,5,7-cyclooctatetraene to [(C₅Me₅)(C₈H ₈)]₂-(C₈H₈) and an eight-electron reduction of 2 equiv of PhN=NPh to form 2 equiv of the U⁶⁺ imido complex (C₅Me₅)₂U(=NPh)₂. In each reaction, H₂ is a byproduct. This hydride-based reduction is also successful with the tetravalent thorium hydride [(C₅Me ₅)₂THH₂]₂ which reduces 2 equiv of PhSSPh to (C₅Me₅)₂-Th(SPh)₂ and 3 equiv of C₈H₈ to [(C₅Me₅)(C ₈H₈)Th]₂(C₈H₈) with concomitant formation of H₂. X-ray crystallographic data are reported on [(C₅Me₅)₂UH]₂, [(C ₅Me₅)₂UH₂]₂, and (C ₅Me₅)₂U(SePh)₂ as well as the thorium reduction products (C₅Me₅)₂Th(SPh) ₂ and [(C₅Me₅)(C₈H ₈)Th]₂(C₈H₈).

Full text of DOI:10.1021/om7003139

3
568
Organometallics 2007, 26, 3568-3576
Actinide Hydride Complexes as Multielectron Reductants:
Analogous Reduction Chemistry from [(C Me ) UH] ,
5
5 2
2
[
(C Me ) UH ] , and [(C Me ) ThH ]
5
5 2
2 2
5
5 2
2 2
,
†
†
†
†
William J. Evans,* Kevin A. Miller, Stosh A. Kozimor, Joseph W. Ziller,
‡
‡
Antonio G. DiPasquale, and Arnold L. Rheingold
Department of Chemistry, UniVersity of California IrVine, California 92697-2025, and Department of
Chemistry and Biochemistry, UniVersity of California, San Diego, 9500 Gilman DriVe, MC 0358,
La Jolla, California 92093-0358
ReceiVed April 4, 2007
Methods to separate the components of the equilibrium mixture of [(C
UH have been developed that allow their reductive chemistry to be studied. These actinide hydrides
can act as four-, six-, and eight-electron reductants depending on the substrate with H as the byproduct
redox couple. This hydride reduction chemistry allows complexes of redox-
Me ThH to be four- and six-electron reductants. [(C Me UH] and [(C
cleanly reduce 2 equiv of PhEEPh (E ) S, Se) to form 2 equiv of (C Me U(SPh) and
U(SePh) in an overall four-electron reduction in each case. [(C Me UH] and [(C Me
also effect a six-electron reduction of 3 equiv of 1,3,5,7-cyclooctatetraene to [(C Me )(C )U]
5 5 2 2 5 5 2
Me ) UH] and [(C Me ) -
2 2
]
2
-
-
1
2 2
of a H f e + / H
inactive Th4 such as [(C
Me UH
Me
UH
+
)
]
)
5
5
2
2
2
5
5
2
2
5
-
5
)
2
]
2 2
5
5
)
2
2
(
C
5
5
)
2
2
5
5
)
2
2
5
5
)
2
-
-
2
]
2
5
+
5
8 8
H
2
6
(C H
8 8
) and an eight-electron reduction of 2 equiv of PhNdNPh to form 2 equiv of the U imido complex
Me U(dNPh) . In each reaction, H is a byproduct. This hydride-based reduction is also successful
Me ThH , which reduces 2 equiv of PhSSPh to (C Me
Me )(C )Th] (C ) with concomitant formation of H . X-ray
crystallographic data are reported on [(C Me UH] , [(C Me UH , and (C Me U(SePh) as well as
Th(SPh) and [(C Me )(C )Th] (C ).
(C
5
5
)
2
2
2
with the tetravalent thorium hydride [(C
Th(SPh) and 3 equiv of C to [(C
5
5
)
8
2
2
]
2
5
5 2
) -
2
8
H
8
5
5
H
8
2
8
H
8
2
5
5
)
2
2
5
5
)
2
2
]
H
2
5
5
)
2
2
the thorium reduction products (C
5
Me
5
)
2
2
5
5
8
8
2
8 8
H
Introduction
Scheme 1
Recent studies of the sterically crowded organoactinide
1
complex (C5Me5)3U have shown that it can function as a
3+
multielectron reductant by combining its traditional U reduc-
tive chemistry2 with (C5Me5) -based reduction. This type
of ligand-based reduction is called sterically induced reduction
since it occurs only in complexes with unusually long metal-
-4
-
5,6
7
ligand bonds. For example, a four-electron reduction can be
3
+
-
effected using 1 equiv of U
and a (C5Me5) /(C5Me5)
5
transformation, Scheme 1. Sterically crowded [(C5Me5)2U]2-
8
6
(
C6H6) reacts similarly.
As a control reaction for this sterically induced reduction
Scheme 2
chemistry, [(C5Me5)2U][(µ-Ph)2BPh2],9 a complex that has
normal bond lengths, was treated with PhNdNPh for compari-
6
son. Surprisingly, the same four-electron reduction shown in
3
+
Scheme 1 occurred as shown in Scheme 2. In this case the U
-
reduction was combined with a (BPh4) -based reduction that
*
Corresponding author. Fax: 949-824-2210. E-mail: wevans@uci.edu.
University of California Irvine.
University of California, San Diego.
†
‡
(
1) Evans, W. J.; Forrestal, K. J.; Ziller, J. W. Angew. Chem., Int. Ed.
1
997, 36, 774.
(
(
(
(
2) Evans, W. J.; Kozimor, S. A. Coord. Chem. ReV. 2006, 250, 911.
3) Castro-Rodriquez, I.; Meyer, K. Chem. Commun. 2006, 1353.
4) Ephritikhine, M. Dalton Trans. 2006, 2501.
did not require steric crowding. Ligand- and metal-based
t
multielectron reduction have also been observed with {[( Bu)-
5) Evans, W. J.; Nyce, G. W.; Ziller, J. W. Angew. Chem., Int. Ed.
1
0
(
C6H3Me2-3,5)N]2U}(C6H5Me), another complex that has
2
000, 39, 240.
6) Evans, W. J.; Kozimor, S. A.; Ziller, J, W. Chem. Commun. 2005,
681.
(
conventional bond distances.
4
To explore the potential for other multielectron reduction
(
7) Evans, W. J. Coord. Chem. ReV. 2000, 206-207, 263.
3
+
reactions using ligand-based processes in U complexes with
(8) Evans, W. J.; Kozimor, S. A.; Ziller, J. W.; Kaltsoyannis, N. J. Am.
Chem. Soc. 2004, 126, 14533.
9) Evans, W. J.; Nyce, G. W.; Forrestal, K. J.; Ziller, J. W. Organo-
metallics 2002, 21, 1055.
(
(10) Diaconescu, P. L.; Arnold, P. L.; Baker, T. A.; Mindiola, D. J.;
Cummins, C. C. J. Am. Chem. Soc. 2000, 122, 6108.
1
0.1021/om7003139 CCC: $37.00 © 2007 American Chemical Society
Publication on Web 05/23/2007

Actinide Hydride Complexes as Multielectron Reductants
Organometallics, Vol. 26, No. 14, 2007 3569
, 2, (C Me U(SePh) , 5, (C Me Th(SPh) , 8,
Table 1. X-ray Data Collection Parameters for [(C
5
Me
5
)
2
UH]
5
2
, 1, [(C
5
Me
(C
5
)
2
UH
2
]
2
5
5
)
2
2
5
5
)
2
2
and [(C
5
Me )(C
8
H
8
)Th]
2
8
H
8
), 9
empirical formula
C40H62U2‚(C7H8)
‚(C7H8)
1111.09
163(2)
monoclinic
P21/n
16.032(2)
14.186(2)
38.211(6)
90
96.765(3)
90
8630(2)
8
1.710
7.526
0.0460
0.1054
C40H64U2
2
1020.99
208(2)
monoclinic
C2/c
18.8970(1)
12.8770(7)
16.2190(9)
90
104.8110(10)
90
3815.5(4)
4
1.774
8.502
C32H40Se2U
5
820.59
100(2)
hexagonal
P61
12.7763(6)
12.7763(6)
32.1519(19)
90
90
120
4545.1(4)
6
1.799
7.777
0.0267
0.0467
C32H40S2Th
8
720.80
168(2)
hexagonal
P65
12.6647(6)
12.6647(6)
32.450(3)
90
90
120
4507.5(5)
6
1.593
C44H54 Th2‚(C6H6)
9‚(C6H6)
1125.06
163(2)
monoclinic
C2
15.372(7)
9.907(4)
13.781(6)
90
103.242(7)
90
2042.9(15)
2
1.829
7.305
0.0273
1
fw
temperature (K)
cryst syst
space group
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
3
volume (Å )
Z
3
Fcalcd (Mg/m )
-
1
µ (mm
)
5.119
0.0235
0.0462
a
R1 [I > 2.0σ(I)]
0.0477
0.1225
a
wR2 (all data)
0.0716
a
2
2 2
2 2 1/2
Definitions: wR2 ) [∑[w(Fo - Fc ) ]/∑[w(Fo ) ] ] , R1 ) ∑||Fo| - |Fc||/∑|Fo|.
normal bond lengths, the reductive chemistry of [(C5Me5)2UH]2,
involved addition of external reductants such as alkali metal
1
6,20-24
1
, was of interest. Hydrides are well-known reductants, but they
naphthalides.
As described here, the tetravalent thorium
11
15,25
generally add to the substrate during the reduction. Since the
hydride, [(C5Me5)2ThH2]2,
3, is also an excellent reductant
f element hydrides are primarily known for insertion and σ bond
and can achieve multielectron reduction reactivity analogous
to that of tetravalent and trivalent uranium hydrides.
metathesis reactivity,1
2-14
a reduction reaction of the type shown
in eq 1 analogous to the ligand-based reductions in Scheme 1
Experimental Section
-
1
-
H f / H + 1e
(1)
2
2
The manipulations described below were performed under argon
with rigorous exclusion of air and water using Schlenk, vacuum
line, and glovebox techniques. Solvents were dried over Q-5 and
and 2 would not be a traditional reaction for these complexes.
In addition, the use of [(C5Me5)2UH]2, 1, was complicated by
the fact that this trivalent hydride typically exists as an
equilibrium mixture with the tetravalent hydride [(C5Me5)2-
2
6
molecular sieves and saturated with argon using GlassContour
columns. Toluene-d and benzene-d were dried over NaK alloy
5 5 2
and vacuum transferred before use. (C (C Me ) -
8
6
15
Me
5
)
2
15
ThCl
2
,
1
5
5
UH2]2, 2, eq 2.
15
15
27
2
UMe ,
[(C
5
Me
5
)
2
2
UH ]
2
,
5 5 2 2 2
[(C Me ) ThH ] , and KSPh were
prepared as previously described. PhSSPh, PhSeSePh, and PhNd
NPh were purchased from Aldrich and sublimed before use. 1,3,5,7-
C H
8 8
was distilled onto molecular sieves before use. NMR spectra
were recorded with a Bruker DRX 500 MHz system. Infrared
spectra were recorded as either thin films obtained from C on
6
H
6
an ASI ReactIR 1000 instrument28 or as KBr pellets on a Perkin-
Elmer Spectrum One FT-IR spectrometer. Elemental analyses were
performed by Analytische Laboratorien, Lindlar, Germany. X-ray
data collection parameters are given in Table 1, and full crystal-
lographic information is in the Supporting Information.
Reported here are conditions in which trivalent 1 can be
isolated from tetravalent 2 such that the reduction chemistry of
each can be examined with several different classes of sub-
strates: PhEEPh (E ) S, Se), 1,3,5,7-cyclooctatetraene, and
PhNdNPh. Encouraged by the reduction chemistry observed
with these organometallic uranium hydrides, extension of this
ligand-based hydride chemistry to thorium was examined.
Reduction chemistry with organometallic thorium complexes
has traditionally been challenging because only three organo-
Crystallization of [(C Me ) UH] , 1. A flask fitted with a high-
5
5 2
2
vacuum greaseless stopcock was charged with a solution of (C5-
Me UMe (100 mg, 0.186 mmol) in benzene (15 mL). The flask
was attached to a high-vacuum line (1 × 10 Torr) and degassed
by three freeze-pump-thaw cycles. H (1 atm) was introduced
5
)
2
2
-
5
2
into the flask at room temperature. The red solution slowly changed
to brown-green as it was stirred for 12 h. The brown-green solid
metallic Th3 compounds have been fully characterized,
+
16
t
17,18
t
(19) Parry, J. S.; Cloke, F. G. N; Coles, S. J.; Hursthouse, M. B. J. Am.
Chem. Soc. 1999, 121, 6867.
[
(RMe2Si)2C5H3]3Th (R ) Me, Bu)
and {[( BuMe2Si)2C8H6]2-
19
Th}K(DME)2. Reduction reactions with thorium have typically
(20) Korobkov, I.; Gambarotta, S.; Yap, G. P. A. Angew. Chem., Int.
Ed. 2003, 42, 814.
(21) Korobkov, I.; Gambarotta, S.; Yap, G. P. A. Angew. Chem., Int.
Ed. 2003, 42, 4958.
(22) Korobkov, I.; Arunachalampillai, A.; Gambarotta, S. Organome-
tallics 2004, 23, 6248.
(23) Korobkov, I.; Gambarotta, S. Organometallics 2004, 23, 5379.
(24) Arunachalampillai, A.; Gambarotta, S.; Korobkov, I. Organome-
tallics 2005, 24, 1996.
(
11) James, B. R. In ComprehensiVe Organometallic Chemistry; Wilkin-
son, G., Gordon, F., Stone, A., Abel, E. W., Eds.; Peragon: Oxford, 1982;
Chapter 51.
(
(
(
(
12) Marks, T. J.; Lin, Z. J. Am. Chem. Soc. 1987, 109, 7979.
13) Marks, T. J.; Lin, Z. J. Am. Chem. Soc. 1990, 112, 5515.
14) Ephritikhine, M. Chem. ReV. 1997, 97, 2193.
15) Marks, T. J.; Fagan, P. J.; Manriquez, J. M.; Maatta, E. A.; Seyam,
A. M. J. Am. Chem. Soc. 1981, 103, 6650.
16) Arunachalampillai, A.; Crewdson, P.; Korobkov, I.; Gambarotta,
S. Organometallics 2006, 25, 3856.
17) Blake, P. C.; Lappert, M. F.; Atwood, J. L.; Zhang, H. J. Chem.
Soc., Chem Commun. 1986, 1148.
18) Blake, P. C.; Edelstein, N. M.; Hitchcock, P. B.; Kot, W. K.; Lappert,
M. F.; Shalimoff, G. V.; Tian, S. J. Organomet. Chem. 2001, 636, 124.
(25) Broach, R. W.; Schultz, A. J.; Williams, J. M.; Brown, G. M.;
Manriquez, J. M.; Fagan, P. J.; Marks, T. J. Science 1979, 203, 172.
(26) For more information on drying systems, see www.glasscontour.com
(27) Evans, W. J.; Miller, K. A.; Lee, D. S.; Ziller, J. W. Inorg. Chem.
2005, 44, 4326.
(28) Evans, W. J.; Johnston, M. A.; Ziller, J. W. Inorg. Chem. 2000, 39,
3421.
(
(
(

3
570 Organometallics, Vol. 26, No. 14, 2007
EVans et al.
was extracted with hexane in an argon-filled glovebox and
centrifuged to remove brown solids. The hexane solution was stirred
for 3 h, the hexane was removed under vacuum, toluene was added,
the solution was stirred for 1 h, and the toluene was removed under
reaction mixture was stirred for 12 h. White insoluble materials
were removed by centrifugation, and solvent was removed under
reduced pressure to yield 8 as a white, crystalline powder (220
mg, 87%). Crystals of 8 suitable for X-ray diffraction were grown
1
1
vacuum, yielding a brown-green powder. H NMR analysis was
at -35 °C from a concentrated toluene solution. H NMR (C
D
6 6
):
3
consistent with the formation of previously reported [(C
5
Me
5
)
2
-
5 5
δ 2.02 (s, 30H, C Me , ∆ν1/2 ) 1 Hz), 6.89 (tt, 2H, JHH ) 7 Hz,
1
5
15
3
3
UH]
2
,
although a small amount of [(C
5
Me
5
)
2
UH
2
]
2
was also
p-H), 7.04 (tt, 4H, JHH ) 8 Hz, m-H), 7.83 (dd, 4H, JHH ) 8 Hz,
1
13
present (15:1 by H NMR spectroscopy). Black polyhedral crystals
suitable for X-ray diffraction were grown at -35 °C after 2 days
from a saturated solution of 1 in toluene.
6 6 5 5 5 5
o-H). C NMR (C D ): δ 12.1 (C Me ), 128.5 (C Me ), 134.0 (o-
phenyl), 129.2 (m-phenyl), 125.3 (p-phenyl), 143.0 (ipso-phenyl).
IR (thin film): 2961s, 2914s, 2856s, 2729w, 2291w, 1579m, 1476m,
1440m, 1378m, 1262s, 1085s, 1065s, 1023s, 803s, 737m, 695s,
Crystallization of [(C
hexane and 4 mL of hot toluene to (C
5
Me
5
)
2
UH
2
]
2
, 2. Addition of 4 mL of hot
Me UMe (668 mg, 1.25
-
1
)
5 2
2
664m cm . Anal. Calcd for C32
8.90; Th, 32.19. Found: C, 53.22; H, 5.57; S, 8.79; Th, 32.05.
(C Me Th(SPh) , 8, from [(C Me ThH , 3. PhSSPh (10
mg, 0.048 mmol) in C was added to a J-Young tube containing
[(C Me ThH , 3 (24 mg, 0.024 mmol), in C . The J-Young
40 2
H S Th: C, 53.32; H, 5.59; S,
5
mmol) produced a saturated red solution that was transferred to a
Fisher Porter high-pressure reaction vessel. This was degassed to
the vapor pressure of the solvent and pressurized to 80 psi with
5
5
)
2
2
5
)
5 2
2 2
]
6 6
D
2
H . After 2 days, large black crystalline rods formed. The pressure
5
5
)
2
2
]
2
6 6
D
was reduced to 20 psi, the reaction vessel was brought into the
argon glovebox, and the vessel was evacuated to the vapor pressure
of the solvent and backfilled with argon (3×). The solvent was
tube was capped immediately, and a color change from pale yellow
to colorless was observed along with the formation of bubbles. H
1
NMR spectroscopy showed quantitative conversion of starting
1
1
decanted to leave black crystals that analyzed by H NMR and IR
material to 8 and H
[(C Me )(C )Th]
mmol) was added dropwise to a pale yellow solution of [(C
ThH , 3 (68 mg, 0.067 mmol), in toluene (10 mL). A dark orange
2
exhibiting a H NMR resonance at 4.46 ppm.
15
spectroscopy to be [(C
Me U(SPh) , 4. PhSSPh (12 mg, 0.055 mmol) in C
added to a J-Young tube containing 1 (29 mg, 0.028 mmol) in C
5
Me
5
)
2
UH
2
]
2
.
5
5
8
H
8
2
(C
8
H
8
), 9. 1,3,5,7-C
H
8 8
(23 µL, 0.203
5 2
Me ) -
(C
5
)
5 2
2
6 6
D was
5
2 2
]
D .
6 6
solution immediately formed. After the mixture was stirred for 3
days, the dark orange solution was evaporated to dryness, yielding
an orange powder. The orange powder was washed with cold
hexane (1 × 5 mL) and dried under reduced pressure to afford 9
(59 mg, 83%). Orange crystals of 9 suitable for X-ray diffraction
The J-Young tube was capped immediately, and a color change
from brown to red was observed along with the formation of
bubbles. H NMR spectroscopy showed quantitative conversion of
1
2
9
starting material to the previously characterized 4 and H
2
1
exhibiting a H NMR resonance at 4.46 ppm.
Me U(SePh) , 5. PhSeSePh (75 mg, 0.240 mmol) in toluene
3 mL) was added to a brown-green solution of 1 (122 mg, 0.120
1
were grown at 25 °C from a concentrated benzene solution. H
(C
5
)
5 2
2
NMR (C
6
D
6
, 25 °C): δ 1.93 (s, 30H, C
, ∆ν1/2 ) 1 Hz). 13C NMR (C
D
6 6
5
Me
5
, ∆ν1/2 ) 1 Hz), 6.15
, 25 °C): δ 13.7
). H NMR (toluene-d , 80
, ∆ν1/2 ) 1 Hz), 6.06 (s, 8H, C
1/2 ) 1 Hz). H NMR (toluene-d , -65 °C): δ 1.91 (s, 30H,
Me , ∆ν1/2 ) 1 Hz), 6.09 (s, 8H, C , ∆ν1/2 ) 1 Hz). IR (thin
film): 2961s, 2918s, 2856s, 1444m, 1378m, 1258vs, 1089vs,
(
(
(
°
s, 8H, C
Me ), 125.9 (C
C): δ 1.92 (s, 30H, C
8
H
8
mmol) in toluene (10 mL). A dark red solution immediately formed
and bubbles were observed. After the mixture was stirred for 12 h,
the dark orange solution was evaporated to dryness, yielding a red
oil. The red oil was dissolved in hexane and cooled to -35 °C.
After 2 days, 5 was obtained as red crystals (132 mg, 67%). Crystals
1
C
5
5
5
Me
5
), 99.21 (C
Me
8
H
8
8
5
5
8 8
H ,
1
∆ν
8
C
5
5
8 8
H
of 5 suitable for X-ray diffraction were grown at -35 °C from a
-
1
1
1085vs, 1015vs, 865s, 803vs, 703s cm . Anal. Calcd for C H -
concentrated hexane solution. H NMR (C
6
D
6
): δ 14.0 (s, 30H
44 54
3
Th : C, 50.48; H, 5.20. Found: C, 51.20; H, 5.43.
C
5
Me
s, 4H, ∆ν1/2 ) 28 Hz, m-H), 0.98 (d, 4H, JHH ) 8 Hz, o-H).
NMR (C ): δ -25.6 (C Me ), 126.0 (C Me ), 104.5 (o-phenyl),
59.7 (m-phenyl), 134.2 (p-phenyl), 129.7 (ipso-phenyl). IR (thin
film): 3053w, 2964m, 2907m, 2856m, 1575m, 1476m, 1436s,
5
, ∆ν1/2 ) 18 Hz), 2.6 (t, 2H, JHH ) 8 Hz, p-H), -31.0 (br
2
3
13
X-ray Data Collection, Structure Determination, and Refine-
C
ment. [(C
sions 0.15 × 0.24 × 0.28 mm was mounted on a glass fiber and
transferred to a Bruker CCD platform diffractometer. The SMART
5 5 2 2
Me ) UH] , 1. A black crystal of approximate dimen-
6
D
6
5
5
5
5
1
31
program package was used to determine the unit-cell parameters
and for data collection (25 s/frame scan time for a sphere of
1
6
378m, 1325w, 1297w, 1262vs, 1089s, 1015vs, 865w, 799s, 773s,
91s, 664s cm . Anal. Calcd for C32H40Se U: C, 46.84; H, 4.91;
2
-
1
3
2
diffraction data). The raw frame data were processed using SAINT
Se, 19.24; U, 29.01. Found: C, 46.88; H, 5.05; Se, 19.41; U, 28.80.
(C Me )(C )U] (C ), 6. 1,3,5,7-C (9.8 µL, 0.087 mmol)
in C was added to a J-Young tube containing [(C Me UH]
(30 mg, 0.029 mmol), in C . The J-Young tube was capped
and SADABS33 to yield the reflection data file. Subsequent
calculations were carried out using the SHELXTL program. The
[
5
5
H
8 8
2
H
8 8
8 8
H
34
D
6 6
5
5
)
2
2
,
diffraction symmetry was 2/m, and the systematic absences were
consistent with the centrosymmetric monoclinic space group P2 /
1
1
6 6
D
immediately, and a color change from brown-green to brown-black
1
n, which was later determined to be correct. The structure was
was observed. H NMR spectroscopy showed quantitative conver-
2
5
solved by direct methods and refined on F by full-matrix least-
sion of starting material to the previously characterized 6, (C
5
-
squares techniques. The analytical scattering factors3 for neutral
atoms were used throughout the analysis. Hydrogen atoms were
included using a riding model. There was one molecule of toluene
solvent present per formula unit. The hydrides could not be located
and were not included in the refinement. At convergence, wR2 )
0.1054 and Goof ) 1.082 for 855 variables refined against 17 628
data (0.80 Å). As a comparison for refinement on F, R1 ) 0.0460
for those 13 375 data with I > 2.0σ(I).
5
Me
(
5
)
C
2
, and H
Me U(dNPh)
was added to a J-Young tube containing [(C
.025 mmol), in C . The J-Young tube was capped immediately,
2
after 3 days.
, 7. PhNdNPh (9 mg, 0.049 mmol) in C
Me UH]
5
)
5 2
2
6
D
6
5
)
5 2
2
, 1 (25 mg,
0
6 6
D
and a color change from brown to brown-red was observed along
1
with the formation of bubbles. H NMR spectroscopy showed
quantitative conversion of starting material to the previously
characterized 76,30 and H
ppm.
1
2
exhibiting a H NMR resonance at 4.46
(
31) SMART Software Users Guide, Version 5.1; Bruker Analytical
X-Ray Systems, Inc.; Madison, WI, 1999.
32) SAINT Software Users Guide, Version 6.0; Bruker Analytical X-Ray
Systems, Inc.; Madison, WI, 1999.
33) Sheldrick, G. M. SADABS, Version 2.10; Bruker Analytical X-Ray
(
C
5
Me
200 mg, 0.35 mmol) in toluene (10 mL) was added to a white
slurry of KSPh (129 mg, 0.87 mmol) in toluene (2 mL), and the
5 2 2 5 5 2 2
) Th(SPh) , 8. A pale yellow solution of (C Me ) ThCl
(
(
(
Systems, Inc.: Madison, WI, 2002.
(34) Sheldrick, G. M. SHELXTL Version 6.12; Bruker Analytical X-Ray
Systems, Inc.; Madison, WI, 2001.
(35) International Tables for X-Ray Crystallography; Kluwer Academic
Publishers: Dordrecht, 1992; Vol. C.
(
29) Lescop, C.; Arliguie, T.; Lance, M.; Nierlich, M.; Ephritikhine, M.
J. Organomet. Chem. 1999, 580, 137.
30) Arney, S. J.; Burns, C. J.; Smith, D. C. J. Am. Chem. Soc. 1992,
14, 10068.
(
1

Actinide Hydride Complexes as Multielectron Reductants
(C Me UH
Organometallics, Vol. 26, No. 14, 2007 3571
[
5
5
)
2
2
]
2
, 2. A black block 0.17 × 0.10 × 0.10 mm in
size was mounted on a Cryoloop with Paratone oil. Data were
collected in a nitrogen gas stream at 208(2) K using phi and omega
scans. The crystal-to-detector distance was 60 mm and exposure
time was 5 s per frame using a scan width of 0.3°. Data collection
was 99.9% complete to 25.00° in θ. A total of 13 055 reflections
were collected covering the indices -25 e h e 13, -16 e k e
17, -21 e l e 19. A total of 4442 reflections were found to be
symmetry independent, with an Rint of 0.0305. Indexing and unit
cell refinement indicated a C-centered, monoclinic lattice. The space
group was found to be C2/c (No. 15). The data were integrated
32
using the Bruker SAINT software program and scaled using the
SADABS33 software program. Solution by direct methods (SIR-
97) produced a complete heavy-atom phasing model consistent with
the proposed structure. All non-hydrogen atoms were refined
anisotropically by full-matrix least-squares (SHELXL-97). All
hydrogen atoms, with the exception of the hydride hydrogen H1u,
were placed using a riding model. Their positions were constrained
relative to their parent atom using the appropriate HFIX command
in SHELXL-97. The hydride hydrogen H1u was located from the
difference map, and its position was refined isotropically.
Figure 1. Molecular structure of one of the crystallographically
independent molecules in the unit cell of [(C Me ) UH] , 1, with
thermal ellipsoids drawn at the 50% level.
5
5 2
2
2
, is the dominant species under H2. However in the absence
of hydrogen, 2 loses H2 to form trivalent 1. To avoid using
mixtures of 1 and 2, attempts were made to isolate each and
confirm their identity by X-ray crystallography.
(
C
5
Me
5
)
2
U(SePh)
2
, 5. A red plate 0.10 × 0.10 × 0.05 mm in
size was mounted on a Cryoloop with Paratone oil. Data were
collected in a nitrogen gas stream at 100(2) K using phi and omega
scans. A total of 19 076 reflections were collected covering the
indices -17 e h e 17, -17 e k e 17, -42 e l e 42. A total of
385 reflections were found to be symmetry independent, with an
int of 0.0366. Indexing and unit cell refinement indicated a
hexagonal lattice. The space group was found to be P6(1) (No.
69). The data were handled as described for 2. The crystal was a
perfect merohedral twin, with a refined BASF parameter of 0.50147-
69).
[(C5Me5)2UH]2, 1. The conversion of 2 to 1 is facilitated by
repeatedly dissolving 2 under argon and removing solvent under
vacuum. Hence, removal of solvent from a toluene solution of
2
followed by dissolution in hexane, removal of solvent, and
6
R
finally dissolution in toluene and removal of solvent generates
1
a brown-green solid that analyzes by H NMR spectroscopy in
-
C6D6 as a 15:1 mixture of 1:2 on the basis of (C5Me5)
1
resonances. X-ray quality crystals were repeatedly grown from
saturated solutions of toluene at -35 °C under argon, and a
(
1
5
structure was obtained. In contrast to the IR spectrum of 2,
5 5 2 2
(C Me ) Th(SPh) , 8. A colorless crystal of approximate dimen-
-
1
which displays absorptions at 1335 and 1180 cm assigned to
a terminal U-H stretching mode and a U-H-U stretching
mode, respectively, these crystals displayed only a single broad
sions 0.12 × 0.12 × 0.21 mm was handled as described for 1. The
diffraction symmetry was 6/m, and the systematic absences were
consistent with the hexagonal space groups P6
1
and P6
5
. It was
-1
band at 1163 cm assignable to a bridging U-H-U stretching
later determined that the correct space group was P6
5
. The structure
2
mode.
was solved by direct methods and refined on F by full-matrix least-
squares techniques. At convergence, wR2 ) 0.0462 and GOF )
An X-ray crystallographic study revealed a structure contain-
ing two similar crystallographically independent molecules in
the unit cell, each of which is comprised of two metallocene
units as shown in Figure 1.
No hydride ligands were located, but the two metallocenes
are presumably bridged by the hydride ligands, as was found
in the neutron diffraction study of tetravalent [(C5Me5)2ThH-
1.102 for 326 variables refined against 7192 data. As a comparison
for refinement on F, R1 ) 0.0235 for those 6673 data with I >
2
.0σ(I). The absolute structure was assigned by refinement of the
36
Flack parameter.
[
(C
dimensions 0.13 × 0.17 × 0.28 mm was handled as described for
. The diffraction symmetry was 2/m, and the systematic absences
5 5 8 8 2 8 8
Me )(C H )Th] (C H ), 9. An orange crystal of approximate
2
5
(
µ-H)]2. The bimetallic structure is equivalent to that found
1
3
7
for [(C5Me5)2SmH]2, 10, another complex in which only the
metallocene components were identified by X-ray crystal-
lography. Complex 1, which crystallized with toluene in the
lattice, is not isomorphous with 10, which crystallized free of
solvents.
were consistent with the monoclinic space group C2, Cm, or C2/
m. It was later determined that the noncentrosymmetric space group
C2 was correct. The structure was solved using the coordinates of
5
2
the uranium analogue and refined on F by full-matrix least-squares
techniques. The molecule was located about a 2-fold rotation axis.
There was one molecule of benzene solvent present per formula
unit. The solvent molecule was also located about a 2-fold axis.
At convergence, wR2 ) 0.0716 and Goof ) 1.034 for 242 variables
refined against 4807 data (0.76 Å). As a comparison for refinement
on F, R1 ) 0.0273 for those 4489 data with I > 2.0σ(I). The
absolute structure was assigned by refinement of the Flack
parameter.36
-
The spatial arrangement of the four (C5Me5) ring centroids
around the U‚‚‚U core can be described as a distorted tetrahe-
dron. As shown in Figure 2, the U(2) ion is almost contained
within the plane defined by its two ring centroids and U(1).
The displacement of this atom is only 0.04 Å out of this plane.
The analogous displacement for U(1) is larger at 0.14 Å. A
similar tetrahedral arrangement was observed in 10, but with
3
7
even less distortion. In 1, the 90(8)° average (ring centroid)-
Results
U-U-(ring centroid) torsional angle is indistinguishable from
3
7
the 90(4)° analogous average in 10. However, the eight
analogous (ring centroid)-U-U-(ring centroid) angles for 1
span a wider range, 81.1° to 104.7°, vs 86.7° to 93.3° in 10.
Each crystallographically independent molecule of 1 has one
Synthesis and Structural Characterization of Uranium
Hydrides. The uranium hydrides [(C5Me5)2UH]2, 1, and [(C5-
Me5)2UH2]2, 2, were originally synthesized by Marks and co-
workers from a reaction between (C5Me5)2UMe2 and H2 that
generated a mixture of 1 and 2, eq 2.15 The tetravalent hydride,
128.5-128.9° (C5Me5 ring centroid)-U-(C5Me5 ring centroid)
(
37) Evans, W. J.; Bloom, I.; Hunter, W. E.; Atwood, J. L. J. Am. Chem.
(36) Flack, H. D. Acta. Crystallogr., Sect. A 1983, 39, 876-881.
Soc. 1983, 105, 1401.

3572 Organometallics, Vol. 26, No. 14, 2007
EVans et al.
5 5 2 2
Figure 2. Ball and stick representation of [(C Me ) UH] , 1.
Figure 3. Molecular structure of [(C
5
Me
5
)
2 2
UH ]
2
, 2, with thermal
Table 2. Selected Bond Distances (Å) and Angles (deg) for
ellipsoids drawn at the 30% level.
[
(C Me UH] , 1
5
5
)
2
2
U(1)-U(2)
U(1)-C(1)
3.8651(7)
2.733(7)
2.762(8)
2.791(8)
2.759(8)
2.736(7)
2.748(8)
2.801(8)
2.803(7)
2.768(7)
2.759(8)
2.482
U(3)-U(2)
3.8530(7)
2.718(8)
2.735(9)
2.740(9)
2.751(9)
2.764(8)
2.747(7)
2.739(8)
2.775(8)
2.811(8)
2.769(8)
2.469
amounts of 2, it is possible that the crystal described above was
the tetravalent uranium hydride 2 and not 1. To clarify this
situation, attempts to get crystallographic data on 2 were made.
When the precursor to 1 and 2, (C5Me5)2UMe2, was placed
under 80 psi of H2, large crystalline rods could be repeatedly
U(2)-C(21)
U(2)-C(22)
U(2)-C(23)
U(2)-C(24)
U(2)-C(25)
U(2)-C(31)
U(2)-C(32)
U(2)-C(33)
U(2)-C(34)
U(2)-C(35)
U(2)-Cnt3
U(1)-C(2)
U(1)-C(3)
U(1)-C(4)
U(1)-C(5)
U(1)-C(11)
U(1)-C(12)
U(1)-C(13)
U(1)-C(14)
U(1)-C(15)
U(1)-Cnt1
U(1)-Cnt2
U(3)-Cnt5
U(3)-Cnt6
Cnt1-U(1)-Cnt2
Cnt5-U(3)-Cnt6
1
obtained. The H NMR spectrum of these crystals matches that
1
5
reported for [(C5Me5)2UH2]2, 2, but over time the spectra
showed conversion to [(C5Me5)2UH]2, 1.
Crystals grown by this procedure diffract X-rays and behave
much differently than the crystals of 1. Repeated attempts to
obtain X-ray diffraction data in the typical low-temperature
range around 165 K were unsuccessful since the crystals cracked
during the data collection. At room temperature, the crystals
showed no sign of cracking in the oil used to mount the crystals.
Data collection at 238 K avoided the decomposition, but gave
2.504
2.488
2.497
128.5
U(2)-Cnt4
2.495
2.495
2.489
133.5
U(4)-Cnt7
U(4)-Cnt8
Cnt3-U(2)-Cnt4
Cnt7-U(4)-Cnt8
128.9
133.8
angle, with the other at 133.5-133.8°. These are in the range
typical for f element metallocenes, but the asymmetry is unusual.
As shown in Table 2, the U-C(C5Me5) bond distances for the
two unique molecules in the unit cell have values between 2.718-
8) and 2.813(8) Å and average 2.76(2) Å. This average is the
same as the 2.76(3) Å analogue in [(C5Me5)2U(µ-Cl)]3, an
eight-coordinate U metallocene. The four U-(C5Me5 ring
centroid) distances in 1 are all in the narrow range 2.488-2.497
Å.
-
data with disorder in the (C5Me5) rings. Ultimately, data
collected at 208 K on a modern instrument that had a shorter
data collection time provided the structure shown in Figure 3.
The complex obtained under hydrogen pressure crystallized
in the space group C2/c, in contrast to the P21/n structure of 1.
Like the structure of 1, the data on 2 show two metallocene
units. In this case, there is only one crystallographically
independent molecule in the unit cell and the two metallocenes
are equivalent by symmetry.
(
3
8
3
+
The U‚‚‚U distances for the two unique molecules in 1 are
The U‚‚‚U distance refined to be 3.606(6) Å, Table 3, a
distance significantly shorter than the 3.8530(7) and 3.8651(7)
Å distances in 1. This is reasonable considering that the Shannon
3
.8530(7) and 3.8651(7) Å. In comparison, the Sm‚‚‚Sm
3
+
distance in 10 is 3.904 Å. Since the Shannon radius for U is
3+
0
.067 Å larger than that of Sm , the U‚‚‚U distance in 1 would
4+
3+
radius for U is 0.135 Å shorter than that for U . The U‚‚‚U
be expected to be longer. In another similar pair of bridged,
eight-coordinate uranium and samarium molecules, the [(C5-
Me5)2M(µ-Cl)]3 complexes, the U‚‚‚U distances averaged 5.669-
distance in 2 is also shorter than the 3.632(2) Å intermetallic
distance in the Th hydride [Me2Si(C5Me4)2ThH2]2. This
difference is in the direction expected on the basis of the nine-
4+
42
(
2) Å compared to 5.633(2) Å for Sm‚‚‚Sm.39 This is in closer
4+
4+
43
coordinate ionic radii of Th , 1.09 Å, and U , 1.05 Å. All
agreement to the Shannon radii. The origin of the difference in
M‚‚‚M is not clear, but this is not the first example of a uranium
bimetallic complex that differs structurally from a composi-
tionally analogous lanthanide complex. For example, the
structure of [(C5Me5)2U]2(µ-O)40 differs from that of the [(C5-
Me5)2Ln]2(µ-O) series (Ln ) La, Nd, Sm, Y)41 in that the
uranium complex contained a 171.5(6)° U-O-U linkage, while
the lanthanide complexes had linear, 180° Ln-O-Ln angles.
of these An‚‚‚An distances are substantially shorter than the
4
.007(8) Å Th‚‚‚Th distance in [(C5Me5)2ThH(µ-Η)]2, structur-
25
ally characterized by neutron diffraction, but this discrepancy
42
has previously been noted. The U-C(C5Me5) distances in 2
range from 2.729(6) to 2.888(6) Å, a wide range that may reflect
the higher than usual data collection temperature. The average,
2
.79(5) Å, is similar to that in 1 within error limits, but the
3+
4+
ranges of U-C(C5Me5) distances for U and U metallocenes
typically overlap. Despite the complication in the data col-
lection and disorder in the (C5Me5) ligands, bridging hydrides
[
(C5Me5)2UH2]2, 2. Since 1 is in equilibrium with 2 and since
44
1
the H NMR spectra of samples of 1 in solution contain trace
-
were located in a difference map and their positions refined
isotropically. No terminal hydrides were located.
(
38) Manriquez, J. M.; Fagan, P. J.; Marks, T. J. J. Am. Chem. Soc.
979, 101, 5075.
39) Evans, W. J.; Drummond, D. K.; Grate, J. W.; Zhang, H.; Atwood,
J. L. J. Am. Chem. Soc. 1987, 109, 3928.
40) Evans, W. J.; Kozimor, S. A.; Ziller, J. W. Polyhedron 2004, 23,
689.
41) Evans, W. J.; Davis, B. L.; Nyce, G. W.; Perotti, J. M.; Ziller, J.
W. J. Organomet. Chem. 2003, 677, 89.
1
(
(42) Fendrick, C. M.; Schertz, L. D.; Day, V. W.; Marks, T. J.
Organometallics 1988, 7, 1828.
(43) Shannon, R. D. Acta Crystallogr., Sect. A 1976, 23, 751.
(44) Evans, W. J.; Miller, K. A.; Ziller, J. W.; Greaves, J. Manuscript in
preparation.
(
2
(

Actinide Hydride Complexes as Multielectron Reductants
Organometallics, Vol. 26, No. 14, 2007 3573
Table 3. Selected Bond Distances (Å) and Angles (deg) for
5 5 2 2 2
[(C Me ) UH ] , 2
U(1)-U(1A)
U(1)-Cnt1
U(1)-Cnt2
3.606(6)
2.499
2.475
U(1)-C(11A)
U(1)-C(12A)
U(1)-C(13A)
U(1)-C(14A)
U(1)-C(15A)
U(1)-C(11B)
U(1)-C(12B)
U(1)-C(13B)
U(1)-C(14B)
U(1)-C(15B)
U(1)-H1U
2.867(6)
2.809(7)
2.794(7)
2.843(7)
2.888(6)
2.755(6)
2.754(7)
2.738(7)
2.729(6)
2.739(7)
1.94(9)
127.6
Cnt1-U(1)-Cnt2
Cnt1-U(1)-H1U
115.3
Cnt1-U(1)-H1UA
Cnt2-U(1)-H1U
Cnt2-U(1)-H1UA
92.6
101.0
5 5 2 2
Figure 4. Molecular structure of (C Me ) U(SePh) , 5, with
thermal ellipsoids drawn at the 50% probability level.
139.5
Scheme 3
Scheme 4
fication of 5 was obtained by X-ray crystallography, Figure 4.
Again, this is a four-electron reduction, Scheme 3.
The structural data that are available as well as the NMR
and IR data on the crystals discussed above suggest that these
are appropriately assigned as 1 and 2. It is possible that both
crystalline forms are two varieties of just one of these
compounds, and it is also possible that the crystals involve
mixtures of the two compounds.45 However, in terms of their
reaction chemistry described next, these crystalline materials
behave as 1 and 2 as assigned.
[(C5Me5)(C8H8)U]2(C8H8), 6. Complex 1 reduces 3 equiv
of 1,3,5,7-C8H8 to quantitatively form the previously reported
5
brown crystalline complex [(C5Me5)(C8H8)U]2(C8H8), 5. The
1
byproducts observed by H NMR spectroscopy were H and
2
(C5Me5)2, Scheme 4. This transformation requires that [(C5-
Me5)2UH]2 deliver six electrons. These can be accounted for
as shown in Scheme 4. Two electrons arise from two U3 ions
and two electrons are generated from the two hydrides, as shown
in Scheme 3. However, in this reaction, two additional electrons
+
Reductive Chemistry of [(C5Me5)2UH]2, 1. (C5Me5)2U-
(SPh)2, 4. Addition of trivalent [(C5Me5)2UH]2, 1, to 2 equiv
of PhSSPh in C6D6 in a J-Young NMR tube caused an
immediate brown to red color change and the evolution of gas.
-
arise from two (C5Me5) ions, presumably from sterically
1
induced reduction involving a crowded intermediate. Sterically
induced reduction has been involved in each previous synthesis
H NMR spectroscopy revealed quantitative formation of the
29
-
previously reported (C5Me5)2U(SPh)2, 4, which has a (C5Me5)
5
,8
1
1
of 6, eqs 4 and 5. The first and second reduction potentials
H NMR resonance at 13.2 ppm. A H NMR resonance at 4.46
4
7
of C8H8 are -1.85 and -1.9 V vs SCE.
C5Me5)2U(dNPh)2, 7. In an attempt to determine if the
reducing power of the hydride ligands in 1 could be combined
ppm consistent with H2 was also observed. Overall, this is a
four-electron reduction, with two electrons formally from the
(
3
+
-
U
ions in 1 and two electrons from the two H ligands,
3
+
6+
with a U to U redox change, 1 was added to 2 equiv of
PhNdNPh. After 12 h, quantitative formation of the previously
Scheme 3. PhSSPh has a reduction potential of -1.6 V vs Ag/
AgNO3.46
6
+
6,30
characterized U bis(imido) complex (C5Me5)2U(dNPh)2,
, was observed along with H2, Scheme 5. In this case, the
(C5Me5)2U(SePh)2, 5. Complex 1 reacts similarly with 2
7
equiv of PhSeSePh, which has a reduction potential of -0.9 V
4
6
overall reaction involved an eight-electron reduction. As shown
in Scheme 5, each hydride delivers one electron as each U3
provides three.
vs Ag/AgNO3, to form (C5Me5)2U(SePh)2, 5, and H2, eq 3.
Red crystals of 5 were isolated in 67% yield and displayed a
+
-
1
(
C5Me5) H NMR resonance at 14.0 ppm. Definitive identi-
Reductive Chemistry of [(C5Me5)2UH2]2, 2. Examining the
reactivity of the hydride ligands in tetravalent 2 is more
complicated because in solution it converts to trivalent 1 and
hydrogen, eq 2. However, since the molecular weights of 1 and
2
differ by only 2 Da, it is possible to get the stoichiometry
(
45) Parkin, G. Acc. Chem. Res. 1992, 25, 455.
(46) Dessy, R. E.; Weissman, P. M.; Pohl, R. L. J. Am. Chem. Soc. 1966,
8
8, 5117.
(47) de Boer, E. AdV. Organomet. Chem. 1964, 2, 115.

3
574 Organometallics, Vol. 26, No. 14, 2007
EVans et al.
that all of these reductions occurred via trivalent 1. As the
concentration of 1 gets depleted, more 2 could convert to 1 via
eq 1.
Reductive Chemistry of [(C5Me5)2ThH2]2, 3. To examine
reductive tetravalent actinide chemistry without the complication
of the equilibrium between 1 and 2, reduction reactions with
15
the tetravalent thorium hydride 3 were examined. In this case,
there is no evidence of a similar tetravalent to trivalent
1
5
2
conversion. In contrast to organometallic uranium chemistry,
3
+
16-19
examples of Th complexes are very rare.
Initially, the reductive reactivity of [(C5Me5)2ThH2]2, 3, was
examined with PhSSPh. Reaction of 3 with 2 equiv of PhSSPh
in a J-Young tube led to gas evolution and quantitative
-
conversion to a product, 8, that displayed a single (C5Me5)
1
resonance in the H NMR spectrum at 2.02 ppm as well as three
unique phenyl resonances. A resonance at 4.46 ppm consistent
1
with H2 was also observed in the H NMR spectrum. The
identity of 8 was determined by independent synthesis from
(C5Me5)2ThCl2 and KSPh, as shown in eq 7. This is analogous
close to that needed for reduction by either uranium reagent.
Since the synthesis of 2 from (C5Me5)2UMe2 and H2 at 1 atm
typically provides 2 as a 50% mixture with 1, this is the material
that was used to evaluate tetravalent uranium hydride reduction.
With PhSSPh, PhSeSePh, C8H8, and PhNdNPh, the same
products described in Schemes 3-5 and eq 3 above, namely,
2
9
to the synthesis of (C Me ) U(SPh) . Definitive characteriza-
5
5 2
2
tion of 8 was accomplished by X-ray crystallography, Figure
5. Hence, 3 reacts with PhSSPh according to Scheme 6 in a
Scheme 5
4
-7, respectively, were observed in reactions with the mixture
Figure 5. Molecular structure of (C
ellipsoids drawn at the 50% probability level.
5 5 2 2
Me ) Th(SPh) , 8, with thermal
of 1 and 2. Equation 6 shows one example. Operationally, this
reaction analogous to Scheme 3. However, in this case the
reduction is due entirely to the hydride ligands, as shown in
Scheme 6.
Scheme 6
means that either the tetravalent or trivalent uranium hydrides
can be used in these reductions. Mechanistically, it is possible

Actinide Hydride Complexes as Multielectron Reductants
Organometallics, Vol. 26, No. 14, 2007 3575
for eight-coordinate U4 and Th . The difference in actinide-
chalcogen bonds, An-E, also matches the differences in the
radii of the components. With the selenium distances expected
to be 0.14 Å longer and uranium lengths 0.05 Å shorter, the
U-Se distance should be 0.09 Å longer than the Th-S distance
on the basis of Shannon radii. Comparison of the 2.801(1) and
+
4+ 43
2
.780(1) Å U-Se distances in 5 and the 2.749(1) and 2.745(1)
Å Th-S distances in 8 gives the expected trend, but the 0.05
Å difference is not as large. If this represents any real difference
in An-E bonding interactions, it is not reflected in the An-
E-C(ipso) angles, which fall in the narrow range 112.7(2)-
1
14.8(2)°. In the (C5Me5)2Sm(EPh)(THF) series, the Sm-E-
C(ipso) angles are also similar: 120.82(17)° for E ) S and
1
the E-C(Ph) distances in 5 and 8 also matches the 0.14 Å
difference in sulfur and selenium size.
2
7
18.51(14)° for E ) Se. The 0.137-0.157 Å difference in
5 5 8 8 2 8 8
Figure 6. Molecular structure of [(C Me )(C H )Th] (C H ), 9,
with thermal ellipsoids drawn at the 50% probability level.
The structure of [(C5Me5)(C8H8)Th]2(C8H8), 9, is isomor-
phous with that of [(C5Me5)(C8H8)U]2(C8H8), 6. As shown in
Table 5, the actinide carbon distances are larger in 9, as expected
for the larger metal, thorium. The structure of 6 was unusual in
Scheme 7
2
-
2-
that it contains a nonplanar (C8H8) dianion. This (C8H8)
bridges the two uranium atoms with a pseudo-allyl orientation
to each. These two allyl units show a common atom, C(11).
Interestingly, this is not a unique structure for uranium, but it
is found also in the thorium structure. Although An-C(12) is
longer for thorium than for uranium as expected, the An-C(11)
and An-C(13) distances are shorter for thorium. The uranium
complex 6 also has a broader range of An-C(pseudo-allyl)
distances, 2.687(1)-3.00(2) Å, compared to thorium, 2.731-
(6)-2.933(6) Å. The consequences of these structural features
on reactivity are under investigation.
Table 4. Selected Bond Distance (Å) and Angles (deg) for
(C Me
5
5
)
2
U(SePh)
2
5
, 5, and (C Me
5
)
2
Th(SPh)
2
, 8
5
8
M ) U; E ) Se
M ) Th; E ) S
Discussion
M(1)-E(1)
M(1)-E(2)
M(1)-Cnt1
M(1)-Cnt2
E(1)-C(21)
E(2)-C(27)
Cnt1-M(1)-E(1)
Cnt1-M(1)-E(2)
Cnt2-M(1)-E(1)
Cnt2-M(1)-E(2)
Cnt1-M(1)-Cnt2
C(21)-E(1)-M(1)
C(27)-E(2)-M(1)
E(1)-M(1)-E(2)
2.8011(7)
2.7997(7)
2.459
2.7488(11)
2.7451(10)
2.525
Trivalent Uranium Hydrides as Multielectron Reductants.
The reduction chemistry in Schemes 3-5 and eq 3 shows that
the hydride ligands in solutions containing a high proportion
of the sterically normal complex [(C5Me5)2UH]2, 1, can function
as effective reducing agents and combine with U3 to effect
multielectron reduction reactions. Solutions of bimetallic 1 have
been shown to accomplish four-, six-, and eight-electron
reductions depending on the substrate. In each case, H2 is a
byproduct. Hence, the hydride ligand reacts cleanly to deliver
an electron according to eq 1 without complications of insertion
or σ bond metathesis chemistry. With the PhSSPh and PhSeSePh
substrates, Scheme 3 and eq 3, the four-electron reductions
involve a simple combination of two H /H and two U /U
redox couples. In the case of the six-electron reduction of 3
equiv of C8H8, Scheme 4, these hydride- and U -based
reductions appear to combine with sterically induced reduction
involving two (C5Me5) ligands since (C5Me5)2 is a byproduct.
2.463
2.525
1.926(6)
1.911(6)
112.8
95.5
94.3
111.1
137.3
112.74(17)
114.8(2)
101.77(2)
1.769(4)
1.774(4)
111.3
95.0
95.8
113.0
135.5
113.76(13)
114.74(14)
103.00(3)
+
[(C5Me5)(C8H8)Th]2(C8H8), 9. In an attempt to make an
analogue of 6, complex 3 was reacted with 3 equiv of C8H8.
Over a period of 3 days, light orange crystalline [(C5Me5)(C8H8)-
Th]2(C8H8), 9, was formed in >80% yield. This product was
fully characterized by X-ray crystallography, Figure 6. The H
NMR spectrum of 9 showed only two resonances at 1.93 and
-
3+ 4+
3+
1
-
6
.15 ppm compared to three resonances at 5.5, -41.7, and -42.2
The eight-electron reduction in Scheme 5 shows that the
hydride-based reduction can also combine with U to U
ppm in paramagnetic 6. Only a single C8H8 resonance was
observed for 9 down to -65 °C. In this case, a six-electron
reduction was accomplished by combining hydride reduction
3+
6+
redox processes.
Tetravalent Uranium Hydrides. The small amount of
tetravalent [(C5Me5)2UH2]2, 2, contained in the solutions of 1
due to the equilibrium in eq 2 did not affect the success of the
reductions above. In fact, it appears that 2 is an equally effective
reductant. Hence, the 1:1 samples of 1 and 2 also effect the
reductions described above. Although each of the reactions done
with these mixtures could proceed through 1, the reductive
capacity of the tetravalent thorium analogue, [(C5Me5)2ThH2]2,
-
with (C5Me5) reduction, presumably from a sterically crowded
intermediate, Scheme 7. This is supported by the fact that (C5-
Me5)2, the signature byproduct of sterically induced reduction,
was observed in the H NMR spectrum. This is the first evidence
of sterically induced reduction chemistry with thorium.
Structural Studies of Reduction Products. The X-ray
crystal structures of (C5Me5)2U(SePh)2, 5, and (C5Me5)2Th-
SPh)2, 8, display conventional metallocene bonding parameters
for formally eight-coordinate tetravalent complexes of this type
as shown in Table 4. The Th-(C5Me5 ring centroid) distances
are approximately 0.05 Å longer than the uranium values,
consistent with the 0.05 Å difference in the Shannon ionic radii
1
4
8
(
3
, suggests that the hydride ligands in 2 could also be one-
electron reductants. In this sense, the four hydride ligands in 2
(
48) Evans, W. J.; Nyce, G. W.; Ziller, J. W. Organometallics 2001, 20,
5489.

3
576 Organometallics, Vol. 26, No. 14, 2007
EVans et al.
(C
), 65
Table 5. Selected Bond Distances (Å) and Angles (deg) for [(C
5
Me
5 8
)(C H
8
)Th]
2
8
(C H
8 5
), 9, and [(C Me
5 8
)(C H
8
)U]
2
8 8
H
[(C5Me5)(C8H8)Th]2(C8H8)
[(C5Me5)(C8H8)U]2(C8H8)
9
6
M(1)-C(11)
2.8495(11)
2.731(6)
2.933(6)
2.575
2.878(2)
2.687(17)
3.00(2)
M(1)-C(12)
M(1)-C(13)
M(1)-Cnt1(C5Me5)
M(1)-Cnt2(C8H8)
2.552
2.058
1.984
(
(
(
(
(
(
(
C5Me5)Cnt1-M(1)-C(11)
C5Me5)Cnt1-M(1)-C(12)
C5Me5)Cnt1-M(1)-C(13)
C8H8)Cnt2-M(1)-C(11)
C8H8)Cnt2-M(1)-C(12)
C8H8)Cnt2-M(1)-C(13)
C5Me5)Cnt1-M(1)-Cnt2(C8H8)
98.6
90.3
99.7
92.4
102.9
104.6
123.6
119.8
137.6
133.1
119
116.5
132.1
134.5
C(11)-M(1)-C(13)
C(12)-M(1)-C(13)
C(12)-M(1)-C(11)
M(1)′-C(11)-M(1)
M-C(C5Me5) range
M-C(C8H8) range
53.30(15)
28.63(16)
29.35(14)
168.4(2)
2.813(6)-2.864(6)
2.702(6)-2.777(6)
53.3(6)
28.4(6)
29.2(6)
172.1(12)
2.782(16)-2.811(18)
2.62(2)-2.723(19)
provide reduction capacity equivalent to the two hydrides and
the two U ions in 1.
Tetravalent Thorium Hydrides. Schemes 6 and 7 show that
versions of organoactinide-based reduction reactions, it is easier
to envisage cycles involving H2 rather than alkali metals as the
stoichiometric reductant.
3
+
[
(C5Me5)2ThH2]2, 3, can function as a four- and six-electron
In general, it appears that any time a synthetic route to a
new actinide complex is needed, hydride complexes should be
considered as precursors. It is unlikely that the actinide hydrides
will cleanly effect reduction and release of H2 in every process,
but certainly the substrates used in this paper demonstrate that
this is possible. The facile reductions with actinide hydrides
observed here suggest that a broader investigation of electro-
positive metal hydride reductions is appropriate. For example,
the trivalent lanthanide hydrides such as [(C5Me5)2LnH]2 may
be equivalently useful as reducing agents.
reductant, respectively, and accomplish reduction analogous to
that of trivalent [(C5Me5)2UH]2, 1, and tetravalent [(C5Me5)2-
UH2]2, 2. The reduction of 2 equiv of PhSSPh to four (SPh)
ligands in Scheme 6 apparently is effected only by hydride
ligands. In Scheme 7, these four hydrides appear to combine
their reduction chemistry with sterically induced reduction to
allow 3 to deliver six electrons overall. It is possible that these
reactions involve a Th intermediate formed by hydride
reduction of Th in analogy with eq 2, but so far no evidence
-
3+
4
+
3+
of a Th intermediate has been observed. Indeed, the paucity
of Th3 complexes
+
16-19
is the reason that this hydride-based
Conclusion
thorium chemistry is so important to this element. The extensive
Efficient ways to make solutions containing a high proportion
of [(C5Me5)2UH]2, 1, have been developed that allow the
reductive chemistry of this trivalent hydride to be defined. This
complex can accomplish four-, six-, and eight-electron reduc-
tions depending on the substrate. The related tetravalent complex
reduction chemistry available through U3
+ 2
has not been
accessible in thorium compounds. Only when external alkali
4+
metal reductants have been added to Th precursors has
reductive chemistry been possible.1
6,20-24
In the case of [(C5-
4
+
Me5)2ThH2]2, 3, the reductants are an integral part of the Th
complex and the byproduct is an easily separated gas, H2.
[(C5Me5)2UH2]2, 2, can effect analogous reductions in which
all of the reduction equivalents arise from the hydride ligands.
Extension of this chemistry to [(C5Me5)2ThH2]2, 3, provides a
method for thorium reduction chemistry to be effected without
The reactions of 3 suggest that much of the reductive trivalent
chemistry of U3 can be extended to thorium for the first time.
+
For example, [(C5Me5)(C8H8)U]2(C8H8), 6, can be formed in
3
+
several ways from U3+ starting materials, e.g., eqs 4 and 5, but
requiring the synthesis of a Th precursor or addition of an
external reductant.
there were no analogous Th3 precursors for [(C5Me5)(C8H8)-
Th]2(C8H8), 9. Since the hydride ligands in 3 are effective
reductants, complexes such as 9 can now be generated for
thorium.
+
More generally, this study shows that the ligand-based
reduction chemistry previously observed with f element com-
-
plexes of (C5Me5) ligands in sterically crowded environments
-
and in (BPh4) compounds can be extended to other ligands
Synthetic Considerations. Since both [(C5Me5)2UH]2, 1, and
common in f element complexes. The hydride case reported
here may just be one example of many to be discovered in the
future.
[(C5Me5)2UH2]2, 2, are synthetically easier to access than the
sterically crowded (C5Me5)3U and [(C5Me5)2U]2(C6H6), com-
plexes 1 and 2 become the reagents of choice to make products
like those shown in Schemes 3-5 and eq 3. The synthesis of
Acknowledgment. We thank the National Science Founda-
tion for support of this research.
[(C5Me5)(C8H8)U]2(C8H8), 6, in particular is now much easier.
This should open up this complex and its thorium analogue for
comparative reactivity studies for the first time.
Supporting Information Available: X-ray diffraction details
Another important difference between 1 and 2 and complexes
like (C5Me5)3U and [(C5Me5)2U]2(C6H6) is that the reduction
equivalents in 1 and 2 are derived from H2 according to eq 2.
The other compounds require alkali metal reduction in their
synthesis. In terms of developing catalytic or at least cyclic
(CIF) and X-ray data collection, structure solution, and refinement
of compounds 1, 2, 5, 6, 8, and 9 (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
OM7003139