Solvent stabilization and hydrogenation catalysis of trimethylphosphine- substituted carbonyl rhenium cations

Article abstract of DOI:10.1021/om034196y

A series of cationic complexes were designed as catalysts for imine hydrogenation processes, and it was anticipated that for this purpose naked 16e^- cations or relatively labile solventcoordinated ones possessing noncoordinating counterions would suffice. Solvento complexes [Re(CO) ₃(PMe₃)₃(S)][BAr^F] (4(PhCl) and 4(THF)) and [mer-Re(CO)₂(PMe₃)₃(S)] [BAr ^F] (5(PhCl) and 5(THF)) (BAr^F = [B(3,5-(CF ₃)₂C₆H₃)₄]^-; S = PhCl)) were obtained from their corresponding hydride complexes 1 and 2 after treatment with [Ph₃C] [BAr^F] in chlorobenzene. The five-coordinated cationic complex [Re(CO)(PMe₃)₄] [BAr^F] (6) (BAr^F = [B(3,5-(CF₃) ₂C₆H₃)₄]-) was obtained by the reaction of ReH(CO)(PMe₃)₄ (3) with 1 equiv of [Ph ₃C][BAr^F] in chlorobenzene. Hydride abstraction also occurred except for 1 from 2 and 3 with B(C₆F₅) ₃ producing [Re(CO)₂(PMe₃)₃(S)] [BH(C₆F₅)₃] and [Re(CO)(PMe₃) ₄] [BH(C₆F₅)] (S = PhCl, THF). Treatment of ReH(CO)₃(PMe₃)₂ (1) and ReH(CO) ₂(PMe₃)₃ (2) with 1 equiv of [isopropylisopropylideneiminium] [BAr^F] in chlorobenzene at room temperature produced a mixture of 4(PhCl) and [Re(CO)₃(PMe ₃)₂(HNⁱPr₂)] [BAr^F] (8) or in the case of 2 a mixture of 5(PhCl) and [Re(CO)₂(PMe ₃)₃(HNⁱPr₂)] [BAr^F] (9) within a few minutes. After 4 h both mixtures were completely converted to 8 and 9, respectively. 8 and 9 could also be obtained reacting 4(PhCl) and 5(PhCl) with excess diisopropylamine. Under mild conditions several imines underwent hydrogenation with H₂ in the presence of 4(PhCl) and 5(PhCl) as catalysts. 6, however, showed only poor catalysis. Further studies revealed details of the mechanism of the catalytic process. X-ray diffraction studies were carried out on the molecular structures of 4(PhCl), 5(PhCl), 6, and 5(THF).

Full text of DOI:10.1021/om034196y

Organometallics 2004, 23, 3153-3163
3153
Solven t Sta biliza tion a n d Hyd r ogen a tion Ca ta lysis of
Tr im eth ylp h osp h in e-Su bstitu ted Ca r bon yl Rh en iu m
Ca tion s
Xiang-Yang Liu, Koushik Venkatesan, Helmut W. Schmalle, and Heinz Berke*
Anorganisch-Chemisches Institut der Universita¨t Zu¨rich, Winterthurerstrasse 190,
CH-8057 Zu¨rich, Switzerland
Received September 25, 2003
A series of cationic complexes were designed as catalysts for imine hydrogenation processes,
and it was anticipated that for this purpose naked 16e- cations or relatively labile solvent-
coordinated ones possessing noncoordinating counterions would suffice. Solvento complexes
[Re(CO)3(PMe3)2(S)][BArF] (4(P h Cl) and 4(THF )) and [mer-Re(CO)2(PMe3)3(S)][BArF] (5-
(P h Cl) and 5(THF )) (BArF ) [B(3,5-(CF3)2C6H3)4]-; S ) PhCl)) were obtained from their
corresponding hydride complexes 1 and 2 after treatment with [Ph3C][BArF] in chlorobenzene.
The five-coordinated cationic complex [Re(CO)(PMe3)4][BArF] (6) (BArF ) [B(3,5-(CF3)2C6H3)4]-)
was obtained by the reaction of ReH(CO)(PMe3)4 (3) with 1 equiv of [Ph3C][BArF] in
chlorobenzene. Hydride abstraction also occurred except for 1 from 2 and 3 with B(C6F5)3,
producing [Re(CO)2(PMe3)3(S)][BH(C6F5)3] and [Re(CO)(PMe3)4][BH(C6F5)] (S ) PhCl, THF).
Treatment of ReH(CO)3(PMe3)2 (1) and ReH(CO)2(PMe3)3 (2) with 1 equiv of [isopropyliso-
propylideneiminium][BArF] in chlorobenzene at room temperature produced a mixture of
4(P h Cl) and [Re(CO)3(PMe3)2(HNiPr2)][BArF] (8) or in the case of 2 a mixture of 5(P h Cl)
and [Re(CO)2(PMe3)3(HNiPr2)][BArF] (9) within a few minutes. After 4 h both mixtures were
completely converted to 8 and 9, respectively. 8 and 9 could also be obtained reacting 4-
(P h Cl) and 5(P h Cl) with excess diisopropylamine. Under mild conditions several imines
underwent hydrogenation with H2 in the presence of 4(P h Cl) and 5(P h Cl) as catalysts. 6,
however, showed only poor catalysis. Further studies revealed details of the mechanism of
the catalytic process. X-ray diffraction studies were carried out on the molecular structures
of 4(P h Cl), 5(P h Cl), 6, and 5(THF ).
In tr od u ction
the activation of H2 by transition-metal species. Experi-
mental and theoretical studies on the interaction of the
H2 molecule with a metal center have provided valuable
insight into how the activation of the H-H bond might
occur.25 The H2 ligand possesses the binding capacities
of both σ-donation and π-back-donation. Depending on
the balance of these parts either homolytic or heterolytic
cleavage of the H2 molecule is preferred. Thus, at an
electron-rich metal complex which is dominated by
d(M)-σ*(H2) back-donation, homolytic cleavage of the
H-H bond is favored.14,26,27 In contrast, at an electron-
deficient metal center σ(H2)-d(M) donation is expected
to dominate, which should preferentially lead to het-
erolytic cleavage of the H-H bond (Scheme 1). Here the
dihydrogen ligand acts as a Bronsted acid.28-31 The pKa
of several H2 ligands in the complexes have been
Homogeneous hydrogenation of organic substrates is
the most widely studied class of catalyzed organome-
tallic reactions.1-24 The key step of these processes is
* Corresponding author. E-mail: hberke@aci.unizh.ch.
(1) Ashby, M. T.; Khan, M. A.; Halpern, J . Organometallics 1991,
10, 2011.
(2) Butts, M. D.; Bergman, R. G. Organometallics 1994, 13, 2668.
(3) Calvin, M. J . Am. Chem. Soc. 1939, 61, 2230.
(4) Chen, J . X.; Daeuble, J . F.; Brestensky, D. M.; Stryker, J . F. M.
Tetrahedron 2000, 56, 2153.
(5) Collman, J . P.; Brauman, J . I.; Tustin, G.; Wann, G. S. J . Am.
Chem. Soc. 1983, 105, 3913.
(6) Coolen, H.; Nolte, R. J . M.; Vanleeuwen, P. J . Organomet. Chem.
1995, 496, 159.
(7) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals; Wiley: New York, 1988.
(8) Halpern, J . J . Organomet. Chem. 1980, 200, 133.
(9) Harmon, R. E.; Gupta, S. K.; Brown, D. J . Chem. Rev. 1973, 73,
21.
(10) Hartmann, R.; Chen, P. Angew. Chem., Int. Ed. 2001, 40, 3581.
(11) Iwasa, T.; Shimada, H.; Takami, A.; Matsuzaka, H.; Ishii, Y.;
Hidai, M. Inorg. Chem. 1999, 38, 2851.
(12) Le, T. X.; Merola, J . S. Organometallics 1993, 12, 3798.
(13) Lee, H. M.; J iang, T.; Stevens, E. D.; Nolan, S. P. Organome-
tallics 2001, 20, 1255.
(14) Osborn, J . A.; J ardine, F. H.; Young, J . F.; Wilkinson, G. J .
Chem. Soc. A 1966, 1711.
(15) Sandee, A. J .; Reek, J . N. H.; Kamer, P. C. J .; van Leeuwen, P.
J . Am. Chem. Soc. 2001, 123, 8468.
(16) Wan, K. T.; Davis, M. E. Tetrahedron: Asymmetry 1993, 4,
2461.
(17) Kubas, G. J . Metal Dihydrogen and σ-Bond Complexes; Kluwer
Academic/Plenum: New York, 2001.
(18) Ephritikhine, M. Chem. Rev. 1997, 97, 2193.
(19) Grushin, V. V. Chem. Rev. 1996, 96, 2011.
(20) J acobsen, H.; Berke, H. Recent Advances in Hydride Chemistry;
Poli, R., Peruzzini, M., Eds.; Elseivier: Amsterdam, 2001.
(21) Berke, H.; Burger, P. Comments Inorg. Chem. 1994, 16, 279.
(22) Heinekey, D. M.; Schomber, B. M.; Radzewich, C. E. J . Am.
Chem. Soc. 1994, 116, 4515.
(23) Heinekey, D. M.; Radzewich, C. E.; Voges, M. H.; Schomber,
B. M. J . Am. Chem. Soc. 1997, 119, 4172.
(24) Huhmann-Vincent, J .; Scott, B. L.; Kubas, G. J . Inorg. Chim.
Acta 1999, 294, 240.
(25) Chalk, A. J .; Halpern, J . J . Am. Chem. Soc. 1959, 81, 5846.
(26) Halpern, J . Adv. Catal. 1959, 11, 301.
(27) Syrkin, Y. K. Usp. Khim. 1959, 28, 903.
10.1021/om034196y CCC: $27.50 © 2004 American Chemical Society
Publication on Web 05/20/2004

3154 Organometallics, Vol. 23, No. 13, 2004
Liu et al.
Sch em e 1
attention to the hydrogenation of imines, as they may
be basic enough to induce heterolytic cleavage of the
H-H bond in the dihydrogen complexes. Therefore, we
decided to further systematically explore the possibility
of a heterolytic activation of dihydrogen by unsaturated
[Re(CO)5-n(PMe3)n]+ cations (n ) 2, 3, and 4) or their
solvent-stabilized analogues. These compounds are for-
mally derived from the corresponding hydrides by
hydride abstraction. It was the first aim of this work to
test whether such cations can be produced by such a
route, and furthermore it was sought to test their
catalytic potential with regard to hydrogenations of
imines.
determined, and the increase in acidity indeed goes
along with a decrease in electron density at the metal
center.32-34 Norton and co-workers have previously
reported the ionic hydrogenation of iminium cations.33
Thus, the pathway for the heterolytic activation of
dihydrogen via deprotonation of H2 complexes is highly
dependent on the ratio of σ-donation/π-acceptor and may
be adjusted by the ancillary ligand sphere. For example,
[cis-Re(CO)4(PR3)(η-H2)]+ is much more acidic than
[mer-Re(CO)3(PR3)2(η-H2)]+, the earlier bearing more of
the π-acceptors CO, allowing its deprotonation by
moderately strong bases. It was also demonstrated by
Kubas and Heinekey that the solvento complexes [Re-
(CO)4(PR3)(S)]+ and [Re(CO)3(PR3)2(S)]+ (S ) CH2Cl2,
Et2O) do react with dihydrogen.22,24,35-39 Recently, our
group has studied the formation and reactivity of such
dihydrogen complexes.40-46 It was found that rhenium
hydrides [Re(CO)5-n(PMe3)nH] (n ) 4, 3, and 2)47,48 react
with strong acids to give dihydrogen complexes [Re-
(CO)5-n(PMe3)n(η-H2)]+ (n ) 3 and 2) at low temperature
and in the case of [Re(CO)5-n(PMe3)nH2]+ (n ) 4) the
dihydride complex.44,45 The hydrides react with acidic
alcohols to give equilibrium mixtures of the hydrides
and the dihydrogen complexes. The dihydrogen species
[Re(CO)5-n(PMe3)n(η-H2)]+ (n ) 4 and 3) derived from
the corresponding hydrides and acidic alcohols allow
stoichiometric hydrogenations of simple imines, alde-
hydes, and ketones. We note that although heterolytic
activation of molecular hydrogen is thought to play a
role in homogeneous hydrogenation reactions, conclusive
experimental evidence is still rare. We paid special
Resu lts a n d Discu ssion
Rea ction s of ReH(CO)3(P Me3)2 (1), ReH(CO)2-
(P Me3)3 (2), a n d ReH(CO)(P Me3)4 (3) w ith [P h 3C]-
[BAr F ] in Ch lor oben zen e. Both ReH(CO)3(PMe3)2 (1)
and ReH(CO)2(PMe3)3 (2) react with 1 equiv of [Ph3C]-
[BArF] in chlorobenzene to give the solvent-coordinated
complexes [Re(CO)3(PMe3)2(S)][BArF], 4(P h Cl), and
[mer-Re(CO)2(PMe3)3(S)][BArF], 5(P h Cl) (BArF ) [B(3,5-
(CF3)2C6H3)4]-; S ) PhCl) (Scheme 2).
The complexes were isolated by crystallization from
PhCl at low temperatures and obtained as yellow
crystalline compounds in high yields. The coordination
of chlorobenzene was confirmed by single-crystal X-ray
diffraction. This contrasts the observation of Heinekey
on a related bisphosphine complex, which was isolated
as a 16e- unsaturated compound after crystallization
from CH2Cl2. Solvent coordination did not occur in this
case. Kubas et al. have reported that related monophos-
phine-substituted cations revealed CH2Cl2 attachment,
suggesting a more electron-deficient and even sterically
less crowded cationic Re center.23,24,35-37 This all shows
that there is a subtle balance of stereoelectronic factors,
which govern solvent coordination. The coordination of
PhCl in 4(P h Cl) and 5(P h Cl) was nevertheless quite
unexpected, since the PhCl molecule is a weaker donor
ligand than CH2Cl2, which was used in the Kubas and
Heinekey experiments. PhCl as a solvent should there-
fore show a higher propensity to produce the “naked”
cations, and furthermore there should be a higher
tendency to yield the naked 16e- species with increasing
amounts of phosphine substituents. Therefore, at least
in the case of the reaction of 2 with [Ph3C][BArF] one
would expect the formation of 5 rather than that of 5-
(P h Cl). There are some literature reports on only a few
examples of η1-PhCl coordination to a metal center,49,50
the structures of which were however not confirmed by
X-ray diffraction studies. The structures of 4(P h Cl) and
5(P h Cl) are shown in Figures 1 and 2. Selected bond
lengths and angles of 4(P h Cl) and 5(P h Cl) are given
in Tables 1 and 2. The coordination geometries of 4-
(P h Cl) and 5(P h Cl) are similar to those of the corre-
sponding hydrides.47 In both complexes PhCl occupies
the original positions of the hydride. The Re-Cl-C
units reveal bending at the chlorine atom, which is
about 113° in both compounds. The Re-Cl distances in
4(P h Cl) and 5(P h Cl) are 2.5613(11) and 2.560(3) Å,
(28) Osakada, K.; Ohshiro, K.; Yamamoto, A. Organometallics 1991,
10, 404.
(29) Grass, V.; Lexa, D.; Saveant, J . M. J . Am. Chem. Soc. 1997,
119, 7526.
(30) Fong, T. P.; Forde, C. E.; Lough, A. J .; Morris, R. H.; Rigo, P.;
Rocchini, E.; Stephan, T. J . Chem. Soc., Dalton Trans. 1999, 4475.
(31) Belkova, N. V.; Bakhmutova, E. V.; Shubina, E. S.; Bianchini,
C.; Peruzzini, M.; Bakhmutov, V. I.; Epstein, L. M. Eur. J . Inorg. Chem.
2000, 2163.
(32) Gusev, D. G.; Hubener, R.; Burger, P.; Orama, O.; Berke, H. J .
Am. Chem. Soc. 1997, 119, 3716.
(33) Magee, M. P.; Norton, J . R. J . Am. Chem. Soc. 2001, 123, 1778.
(34) J ia, G.; Morris, R. H. J . Am. Chem. Soc. 1991, 113, 875.
(35) Huhmann-Vincent, J .; Scott, B. L.; Kubas, G. J . Inorg. Chem.
1999, 38, 115.
(36) Huhmann-Vincent, J .; Scott, B. L.; Kubas, G. J . J . Am. Chem.
Soc. 1998, 120, 6808.
(37) Butts, M. D.; Scott, B. L.; Kubas, G. J . J . Am. Chem. Soc. 1996,
118, 11831.
(38) Heinekey, D. M.; Oldham, W. J . Chem. Rev. 1993, 93, 913.
(39) Heinekey, D. M.; Voges, M. H.; Barnhart, D. M. J . Am. Chem.
Soc. 1996, 118, 10792.
(40) Liu, X. Y.; Bouherour, S.; J acobsen, H.; Schmalle, H. W.; Berke,
H. Inorg. Chim. Acta 2002, 330, 250.
(41) Llamazares, A.; Schmalle, H. W.; Berke, H. Organometallics
2001, 20, 5277.
(42) Messmer, A.; J acobsen, H.; Berke, H. Chem.-Eur. J . 1999, 5,
3341.
(43) Dahlenburg, L.; Herbst, K.; Berke, H. J . Organomet. Chem.
1999, 585, 225.
(44) Feracin, S.; Bu¨rgi, T.; Bakhmutov, V. I.; Eremenko, I.; Voronts-
ov, E. V.; Vimenits, A. B.; Berke, H. Organometallics 1994, 13, 4194.
(45) Gusev, D. G.; Nietlispach, D.; Eremenko, I. L.; Berke, H. Inorg.
Chem. 1993, 32, 3628.
(46) Gusev, D. G.; Kuznetsov, V. F.; Eremenko, I. L.; Berke, H. J .
Am. Chem. Soc. 1993, 115, 5831.
(47) Nietlispach, D.; Veghini, D.; Berke, H. Helv. Chim. Acta 1994,
77, 2197.
(49) Korolev, A. V.; Delpech, F.; Dagorne, S.; Guzei, I. A.; J ordan,
R. F. Organometallics 2001, 20, 3367.
(50) Song, H.; Lee, Y.; Choi, Z. H.; Lee, K.; Park, J . T.; Kwak, J .;
Choi, M. G. Organometallics 2001, 20, 3139.
(48) J acobsen, H.; Berke, H. Chimia 1997, 51, 550.

Carbonyl Rhenium Cations
Organometallics, Vol. 23, No. 13, 2004 3155
F igu r e 2. Molecular structure of 5(P h Cl) (30% probability
displacement ellipsoids). Hydrogen atoms and the BArF
anion are omitted for clarity.
F igu r e 1. Molecular structure of 4(P h Cl) (30% probability
displacement ellipsoids). Hydrogen atoms, chlorobenzene,
and the BArF anion are omitted for clarity.
Ta ble 1. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p lex 4(P h Cl)
respectively, which are comparable to those found for
CH2Cl2 coordination in [Re(CO)4(PiPr3)(CH2Cl2)][BArF],
2.554(2) Å, and [Re(CO)4(PPh3)(CH2Cl2)][BArF], 2.546-
(2) Å.35-37 As expected, the Re-Cl distances of 4(P h Cl)
and 5(P h Cl) are significantly longer than the cor-
responding neutral LnRe-Cl compounds (2.478 Å in
Re(CO)3(PEt3)2Cl51 and 2.511 Å in fac-Re(CO)2(Ph2-
PCH2C(CH3)CH2PPh2CH2PPh2)Cl).52 The Re-C dis-
tances for the CO group trans to PhCl (1.884(4) Å in
4(P h Cl) and 1.896(2) Å in 5(P h Cl)) are definitely
shorter than those for the CO groups trans to a CO or
PMe3 ligand (1.975 Å average in 4(P h Cl)) or CO trans
to PMe3 (1.946 Å in 5(P h Cl)).
Re(1)-Cl(1)
Re(1)-C(2)
Re(1)-P(1)
C(1)-O(1)
C(3)-O(3)
2.5613(11)
1.982(4)
2.4236(12)
1.145(5)
Re(1)-C(1)
Re(1)-C(3)
Re(1)-P(2)
C(2)-O(2)
Cl(1)-C(10)
1.967(4)
1.884(4)
2.4221(11)
1.135(5)
1.763(4)
1.148(5)
P(1)-Re(1)-P(2) 178.57(4)
C(1)-Re(1)-C(2)
176.47(18)
87.22(13)
91.23(4)
Cl(1)-Re(1)-C(3) 173.98(12) Cl(1)-Re(1)-C(1)
Cl(1)-Re(1)-C(2)
Cl(1)-Re(1)-P(2)
95.89(12) Cl(1)-Re(1)-P(1)
88.96(4)
Re(1)-Cl(1)-C(10) 113.18(16)
The spectroscopic data of complexes 4(P h Cl) and 5-
(P h Cl) in solution are consistent with their solid state
structures. The 31P NMR spectrum of 4(P h Cl) shows a
1
singlet at -32 ppm, and the H NMR spectrum exhibits
Sch em e 2

3156 Organometallics, Vol. 23, No. 13, 2004
Liu et al.
Sch em e 3
Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p lex 5(P h Cl)
Ta ble 3. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p lex 6
Re(1)-Cl
2.560(3)
1.896(8)
2.4638(19)
1.128(9)
1.690(9)
Re(1)-C(1)
Re(1)-P(1)
Re(1)-P(3)
C(2)-O(2)
1.947(8)
2.436(2)
2.428(2)
1.153(10)
Re(1)-C (1)
Re(1)-P(21)
Re(1)-P(22)
1.83(4)
2.457(8)
2.377(7)
Re(1)-P(11i)
Re(1)-P(12)
C(1)-O(1)
2.453(6)
2.407(6)
1.12(4)
Re(1)-C(2)
Re(1)-P(2)
C(1)-O(1)
Cl -C(12)
P(11i)-Re(1)-P(12) 106.4(3)
P(12)-Re(1)-P(22) 139.1(3)
P(11i)-Re(1)-P(22) 113.2(3)
C(1)-Re(1)-P(21)
177.5(12)
88.1(12)
94.5(3)
P(1)-Re(1)-P(3) 173.36(8)
C(1)-Re(1)-P(2) 177.0(2)
Cl-Re(1)-C(1)
Cl-Re(1)-P(1)
C(1)-Re(1)-P(11i)
C(1)-Re(1)-P(22)
P(21)-Re(1)-P(12)
86.9(12) C(1)-Re(1)-P(12)
Cl-Re(1)-C(2)
Cl-Re(1)-P(2)
Cl-Re(1)-P(3)
175.5(2)
88.98(8)
89.31(12) Re(1)-Cl-C(12)
94.0(2)
86.91(12)
114.2(4)
84.4(12) P(21)-Re(1)-P(11i)
93.5(3)
P(21)-Re(1)-P(22)
93.2(3)
a broad singlet at 1.21 ppm for the methyl groups. The
phosphorus resonances of the complex 5(P h Cl) in PhCl
are interpreted in terms of a higher order AB2 spectrum
with very close chemical shifts at -35.7 ppm (1P) and
-34.9 ppm (2P). Complex 5(P h Cl) appears to transform
in PhCl in a quite strongly temperature-dependent
equilibrium to other isomers. At temperatures below 0
°C signals for this second species 5a (P h Cl) are ob-
served, which are almost not visible above room tem-
perature. It also exhibits an AB2 pattern in the 31P NMR
with resonances at -43.3 and -44.0 ppm. It is assumed
that its structure corresponds to a fac-phosphine-
substituted cation. The rearrangement process is as-
sumed to involve PhCl dissociation and subsequent
rearrangement of the coordination sphere. Further
support for the structure of 5a (P h Cl) comes from the
fact that there is a great similarity of the 31P NMR
spectra with those of the fac-mer isomerization of the
related hydrides in polar solvents.53
Interestingly, cis-ReH(CO)(PMe3)4 (3) reacts with 1
equiv of [Ph3C][BArF] in chlorobenzene to give the
“naked” 16e- complex [Re(CO)(PMe3)4][BArF] (6) (BArF
) [B(3,5-(CF3)2C6H3)4]-) (Scheme 3). The complex was
isolated as yellow crystals in high yield by crystalliza-
tion from PhCl at low temperature.
The structure of 6 was also confirmed by an X-ray
diffraction study, which revealed a five-coordinate rhe-
nium center (Figure 3). Selected bond lengths and
angles are given in Table 3. The coordination geometry
around the rhenium atom can be best described as a
highly distorted trigonal bipyramid. The P11i, P12, and
P22 atoms occupy the equatorial plane, in which the
main distortion from the ideal geometry is found. Their
arrangement deviates from a triangle in the following
way: while the angle P11i-Re1-P12 is close to 120°,
F igu r e 3. Molecular structure of 6 (30% probability
displacement ellipsoids). Hydrogen atoms and the BArF
anion are omitted for clarity. i ) 1.5-x, y, 1.5-z. 2-axis
symmetry is removed.
the angles P11i-Re1-P22 and P12-Re1-P22 are 106.4-
(3)° versus 139.1(3)°. Thus, a leaning over of P22 toward
P11i with respect to its ideal position of P22 has
occurred, which can be interpreted in terms of a pseudo-
J ahn-Teller distortion.54,55 This change in geometry
affects the doubly degenerate d-orbital set of an ideal
trigonal bipyramid occupied by six electrons having a
triplet ground state. The degeneracies of the d-orbital
sets are lifted, which consequently induces a singlet
ground state with energetic stabilization of the system.
The angles between the axial phosphine ligand (P21)
and the equatorial ones are close to 90° and show
therefore only little distortion with respect to the
(51) Bucknor, S.; Cotton, F. A.; Falvello, L. R.; Reid, A. H.; Schmul-
bach, C. D. Inorg. Chem. 1986, 25, 1021.
(52) Gibson, D. H.; He, H. Y.; Mashuta, M. S. Organometallics 2001,
20, 1456.
(53) Pietsch, J .; Wolski, A.; Dahlenburg, L.; Moll, M.; Berke, H.;
Veghini, D.; Eremenko, I. L. J . Organomet. Chem. 1994, 472, 55.
(54) Albright, T. A.; Burdett, J . K.; Whangbo, M.-H. Orbital Interac-
tions in Chemistry; J ohn Wiley & Sons Ltd: Sussex, 1985.
(55) Burdett, J . K. Chemical Bonds: A Dialog; J ohn Wiley & Sons
Ltd: West Sussex, 1997.

Carbonyl Rhenium Cations
Organometallics, Vol. 23, No. 13, 2004 3157
geometry of an ideal trigonal bipyramid. Presumably
the solvent-deficient five-coordinate structure gains
stabilization not only from the described electronic effect
but also from the fact that a high degree of donor
substitution causes accumulation of electron density on
the metal center, which leads to electronic repulsion of
entering ligands. Furthermore there is steric congestion
of the four phosphine ligands, which apparently do not
leave enough room for the attachment of a just weakly
coordinating ligand. The average of the four Re-P
distances of 6 is 2.424 Å, which is close to the values of
the complexes 4(P h Cl) (2.423 Å) and 5(P h Cl) (2.443
Å). The Re-CO distance in 6 is 1.83(4) Å, which is
shorter than the related average distance in complex
4(P h Cl).
1
The 31P NMR and H NMR spectra of complex 6 in
PhCl display qualitatively a signal pattern similar to
that of the parent cis-hydride 3. Two doublets of triplets
are observed in the 31P NMR appearing at -42.7 and
-47.9 ppm, respectively, together with a doublet of
doublets at -49.1 ppm. Furthermore in the 1H NMR
spectrum a triplet at 1.18 ppm, a doublet at 1.05 ppm,
and another doublet at 0.98 ppm were observed. This
is indicative of a dynamic process moving one of the
equatorial PMe3 groups (P22 in the solid state structure
of 6). It shifts place from one side to the other rapidly
on the NMR time scale, thereby simulating a more
symmetric environment with averaging of the other two
equatorial P nuclei (P11i and P12 in the solid state
structure of 6). An averaging of all three equatorial P
atoms apparently does not take place. It would indeed
require another geometric rearrangement pathway,
which however could be of higher energy.
F igu r e 4. Molecular structure of 5(THF ) (30% probability
displacement ellipsoids). Hydrogen atoms and the BArF
anion are omitted for clarity.
Ta ble 4. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p lex 5(THF )
Re(1)-C(1)
Re(1)-P(3)
Re(1)-P(2)
1.886(6)
2.4161(17)
2.4918(17)
Re(1)-P(1)
Re(1)-O(3)
C(1)-O(1)
2.4430(15)
2.266(4)
1.164(7)
P(3)-Re(1)-P(1) 171.33(6)
O(3)-Re(1)-P(1)
O(3)-Re(1)-P(2)
87.06(12)
95.06(12)
P(3)-Re(1)-P(2)
P(1)-Re(1)-P(2)
C(1)-Re(1)-C(2)
O(3)-Re(1)-P(3)
95.59(7)
90.93(6)
84.5(3)
C(1)-Re(1)-O(3) 177.9(2)
C(2)-Re(1)-O(3)
93.4(2)
93.2(2)
86.68(12) C(1)-Re(1)-P(3)
Rea ction s of ReH(CO)3(P Me3)2 (1), ReH(CO)2-
(P Me3)3 (2), an d cis-ReH(CO)(P Me3)4 (3) with [P h 3C]-
[BAr F ] in THF . Complexes 1, 2, and 3 react with 1
equiv of [Ph3C][BArF] in THF to afford the correspond-
ing solvent-coordinated complexes [Re(CO)3(PMe3)2-
(THF)][BArF], 4(TH F ), [mer-Re(CO)2(PMe3)3(THF)]-
[BArF], 5(THF), and the “naked” species 6. The reactions
in THF are instantaneous and much faster than in
PhCl. The characterization of the new species is based
on their spectroscopic data in comparison with their
analogous complexes 4(P h Cl) and 5(P h Cl). The 31P
NMR spectrum of [Re(CO)3(PMe3)2(THF)][BArF], 4-
(THF ), in THF-d8 appears as a singlet at -28 ppm. The
1H NMR spectrum consists of a singlet at 1.30 ppm for
the methyl groups. Similarly, the NMR spectra of [Re-
(CO)2(PMe3)3(THF)][BArF], 5(THF ), are very close to
those of complex 5(P h Cl); however, from the NMR
spectra there was no indication for a mer-fac rear-
rangement. For instance the 31P NMR spectrum shows
a very related splitting pattern with a triplet at -39.2
ppm and a doublet at -29.5 ppm. The meridional
structure of 5(THF ) was confirmed by an X-ray diffrac-
tion study and is shown in Figure 4. Selected bond
lengths and angles of 5(THF ) are given in Table 4. The
Re-O3 distance in 5(THF ) is 2.266(4) Å. The Re-C
distance for the ReCO group trans to THF is 1.886 Å,
significantly shorter than for the CO group trans to
PMe3 (1.918 Å), reflecting the larger trans-influence of
a PMe3 versus a THF ligand. Therefore, it is reasonable
to assume that these new THF ligating species 4(THF )
and 5(THF ) are structurally related to their chloroben-
zene congeners 4(P h Cl) and 5(P h Cl). It is interesting
to note that 6 does not even coordinate THF despite the
fact that the electron-donating capacity of this molecule
is significantly higher than that of PhCl. This indicates
that the steric repulsion is indeed a major factor for the
existence of 6, which presumably prevents access of
many types of solvent molecules to its rhenium center.
Rea ction s of ReH(CO)2(P Me3)3 (2) a n d ReH(CO)-
(P Me3)4 (3) w it h B(C6F 5)3 in Ch lor oben zen e a n d
THF . While complex 1 does not react with B(C6F5)3,
complexes 2 and 3 are transformed with 1 equiv of
B(C6F5)3 in both chlorobenzene and THF to give the
corresponding hydride-abstracted solvento species [Re-
(CO)2(PMe3)3(S)][BH(C6F5)3] (S ) chlorobezene and
THF) and the complex [Re(CO)(PMe3)4][BH(C6F5)3],
respectively. All the complexes of the type [Re(CO)2-
(PMe3)3(S)][BH(C6F5)3] and [Re(CO)(PMe3)4][BH(C6F5)]
are noncrystalline oily substances, which made their
isolation in pure form difficult. The assignments of their
structures therefore had to be based on spectroscopic
grounds in solution, but could nevertheless safely be
assigned in comparison with the data for complexes 5-
(P h Cl,THF ) and 6. The spectroscopic data of the
cationic parts of these compounds were thus very
similar to those with the [BArF]- counterion and are
therefore not discussed here. The anions however gave
rise to some changes in the1H NMR spectra. The most
important is the appearance of an additional broad
quartet type signal at about 4 ppm, which was at-
tributed to the hydrogen atom of the [BH(C6F5)3]- anion,

3158 Organometallics, Vol. 23, No. 13, 2004
Liu et al.
which has also been identified in [Zr][BH(C6F5)3] com-
plexes by Marks et al.56-61 Due to the noncrystalline
nature of the [BH(C6F5)3]- salts, satisfactory elemental
analysis could not be obtained.
multiplet resonance at about -5.83 ppm split by phos-
phorus coupling.45 Determination of the T1 relaxation
time for the signal of 7a resulted in a value of 12.7 ms
at -30 °C; due to its shortness, it confirms the presence
of a H2 ligand. For the dihydride species 7b the
resonance at -5.83 ppm shows a T1 value of >100 ms,
which is a low value for isolated terminal hydride
P r eca ta lytic In vestiga tion s of th e Solven t-Sta -
bilized Ca tion s 4 a n d 5. (1) Solven t Exch a n ge
Rea ction s of th e Com p lexes 4(solven t) a n d 5(sol-
ven t). Previously in this paper it has been demon-
strated that the cations 4 and 5 can be stabilized with
different solvent donors and different counterions, while
the 16e- cation 6 is stable by itself. Because of the fact
that these complexes were designed as catalysts for
hydrogenation processes, it was anticipated that solvent
lability of 4(P h Cl) and 5(P h Cl) in combination with
the presence of a noncoordinating counterion would even
be a basic prerequisite for their catalytic activity. We
found that THF coordinates much stronger to the
rhenium center than chlorobenzene. For example, the
NMR spectra of 4(THF ) and 5(THF ) dissolved in PhCl-
d5 stay unchanged over several hours. In contrast to this
the PhCl ligand in 4(P h Cl) and 5(P h Cl) is readily
substituted by THF. However, it should be mentioned
at this point that catalysis experiments given in the
later context indicate that the type of coordinated
solvent molecules has no crucial effect on the reactivity
of the Re cations. Nevertheless, solvent stabilization is
principally a suitable form to deliver the cations 4 and
5 as catalyst precursors for reactions with H2.
17,20,30
ligands, but too large a value for a H2 ligand.
Attempts to isolate either the hydride or the dihydrogen
complex led to other yet unidentified species apparently
with loss of the dihydrogen or the hydride moieties.
In a fashion similar to 4(P h Cl) and 5(P h Cl) complex
6 was subjected to 1 bar of H2 in C6D5Cl at -30 °C. It
was found that it reacts with dihydrogen to result in a
complicated mixture of complexes. Among the compo-
nents in the mixture one finds the known dihydride
complex indicated by a multiplet resonance at -6.23
ppm. In this experiment, we could not find a resonance
for the dihydrogen complex, which was expected as a
broad resonance between -6 and -7 ppm.45 This
presumably is due to a stronger preference for the
dihydride species based on the presence of the nonco-
ordination counterion in comparison with the earlier
studies. The dihydride was observed as a characteristic
multiplet at -7.0 ppm with a T1 time of 259.2 ms. The
31P NMR studies revealed a splitting pattern with
multiplets at -26.2, -36.9, and -41.2 ppm. Due to the
coupling pattern of 31P nuclei, we suggest that the
second species is a pentagonal bipyramidal cis dihydride
with the hydride ligands in the equatorial plane. This
presumably is the kinetic product, which appears first
and transforms to the other hydride species. The exist-
ence or coexistence of dihydrogen or dihydride com-
plexes derived from solvent-stabilized 4 and 5 was taken
as a major indication for their potential to act as
precatalysts in hydrogenations of imines. However, the
detection of merely the dihydride complex was taken
as an indication for the poor catalytic performance of 6
because the hydrogenation catalysis obviously requires
the presence of a dihydrogen complex.
(2) Rea ction of 4(P h Cl), 5(P h Cl), a n d 6 w ith H 2.
A [Re(CO)n(PMe3)5-nH2]+ series of complexes bearing
nonclassical H2 or dihydride ligands have been described
earlier by us.45 These compounds were however obtained
by protonation of the corresponding hydride with CF3-
COOH. Thus, the access to these H2 or dihydride
complexes starting from the corresponding cations or
their solvent-stabilized precursors has yet to be verified.
It should furthermore be recognized that the counter-
anions of the earlier and the present experiments are
different, which principally could influence the physical
1
and chemical behavior of the target molecules. In a H
NMR experiment in C6D5Cl 4(P h Cl) was placed under
1 bar of H2. A signal for a dihydrogen complex developed
at -5.70 ppm. This resonance for the H2 ligand pos-
sesses a T1 relaxation time of 13 ms at this temperature,
suggesting its nonclassical type of binding. Related to
this reaction of 4(P h Cl) with H2 is that of a bisphos-
phine rhenium carbonyl cation, which even gives rise
to a thermally stable dihydrogen complex.17,22,23 Com-
plex 5(P h Cl) dissolved in C6D5Cl at -30 °C was also
put under 1 bar of H2, which, somewhat in contrast to
the earlier experiment of protonation of 2,45 afforded an
equilibrium mixture of the dihydrogen complex 7a and
the dihydride complex 7b.
(3) Rea ction of ReH(CO)3(P Me3)2 (1), ReH(CO)2-
(P Me3)3 (2), a n d cis-ReH(CO)(P Me3)4 (3) w ith [Iso-
p r op ylisop r op ylid en eim in iu m ][BAr F ]. The com-
pounds ReH(CO)3(PMe3)2 (1) and ReH(CO)2(PMe3)3 (2)
react readily with 1 equiv of [isopropylisopropylidene-
iminium][BArF] in chlorobenzene at room temperature
to give a mixture of 4(P h Cl) and [Re(CO)3(PMe3)2(HNi-
Pr2)][BArF] (8) or in the case of 2 to yield 5(P h Cl) and
[Re(CO)2(PMe3)3(HNiPr2)][BArF] (9) within a few min-
utes. These insertions of the iminium salts into the
Re-H bond were monitored by 1H and 13C NMR
spectroscopy, which indicated complete conversion to the
amine complexes 8 and 9 after 4 h. After the hydride
transfer step competition between the faster coordina-
tion of chlorobenzene and slower coordination of the
diisopropylamine ligand was noticed. This was further
pursued in separate ligand exchange experiments using
31P NMR spectroscopy. These demonstrated that the
transformations of 4(P h Cl) and of 5(P h Cl) with excess
diisopropylamine to yield 8 and 9, respectively, go
to completion at room temperature. The reaction of
cis-ReH(CO)(PMe3)4 (3) with [Isopropylisopropylidene-
iminium][BArF] is instantaneous and affords at room
temperature the known dihydride complex [ReH2(CO)-
In the 1H NMR spectrum of a PhCl-d5 solution, 7a
was identified by a broad resonance at about -5.76 ppm,
with no recognizable phosphorus coupling, and 7b by a
(56) Beswick, C. L.; Marks, T. J . J . Am. Chem. Soc. 2000, 122, 10358.
(57) Li, L. T.; Stern, C. L.; Marks, T. J . Organometallics 2000, 19,
3332.
(58) Beswick, C. L.; Marks, T. J . Organometallics 1999, 18, 2410.
(59) Deck, P. A.; Beswick, C. L.; Marks, T. J . J . Am. Chem. Soc.
1998, 120, 12167.
(60) Lanza, G.; Fragala, I. L.; Marks, T. J . J . Am. Chem. Soc. 1998,
120, 8257.
(61) Chen, Y. X.; Metz, M. V.; Li, L. T.; Stern, C. L.; Marks, T. J . J .
Am. Chem. Soc. 1998, 120, 6287.

Carbonyl Rhenium Cations
Organometallics, Vol. 23, No. 13, 2004 3159
Ta ble 5. NMR P u r su it of Ca ta lytic Hyd r ogen a tion s of Im in es u n d er H2 Atm osp h er e (1-2 ba r ) Usin g
5(P h Cl) a n d 6 a s Ca ta lysts a t Va r iou s Tem p er a tu r es
cat./imine
(mol %)
temp
(°C)
time
(h)
TOF
yield of
amine (%)
cat.
imine
solvent
C6D5Cl
(h-1
)
5(P h Cl)
PhCHdNCH2Ph
4.0
5.6
2
r.t.
50
90
70
70
90
90
90
90
22
22
70
2
85
85
2
0.23
0.32
0.36
1.70
0.12
0.20
3.5
20
28
100
19
57
93
14
20
30
5(P h Cl)
6
Me2CdNPri
THF/Tol (1:1)
C6D5Cl
PhCHdNCH2Ph
20
150
0.5
0.1
(PMe3)4][BArF] by simple proton transfer in quantitative
yield.45 Concomitantly, isopropylazaisopropylidene was
identified in the reaction mixture by 13C NMR spectros-
copy.
Complexes 8 and 9 possess spectroscopic data very
similar to those of 4(P h Cl) and 5(P h Cl). The presence
of coordinated diisopropylamine could be verified through
1H NMR and 13C NMR spectroscopy. Their structures
are thus consistent with the sketches below.
°C and is even the prevailing species under the given
reaction conditions. The reactions are quite temperature
dependent. While the hydrogenations of PhCHdNCH2-
Ph and Me2CdNiPr with 5(P h Cl) could practically be
brought to completion at about 90 °C, that of PhCHd
NCH2Ph with 6 resulted in lower yields and long
reaction times. 1H NMR and 31P NMR monitoring of
these reactions also revealed that with the progressing
reaction not only does the amine complex accumulate
but also more slowly the free amine appears. This
indicates that the reaction is catalytic and replacement
of the amine ligand by H2 is rate determining. Accord-
ingly, imines with product amines possessing low coor-
dination abilities are even expected to be favorable
substrates in the hydrogenations. This could in particu-
lar be the case for the catalysis of complex 6, which
appears sterically more demanding than the corre-
sponding cationic Re(CO)2(PMe)3 fragment 5. However,
presumably due to its less appropriate acid/base proper-
ties complex 6 was not found to be a very efficient
catalyst, furnishing low amine yields (Table 5) and
catalyst decomposition was noticed by 31P NMR moni-
toring. The slow replacement of the amine ligands by
H2 points to the possibility that these catalyses with
Re(CO)n(PMe3)5-n cations could greatly be improved by
applying H2 under pressure.
These experiments established at least for 4(P h Cl)
and 5(P h Cl) the existence of essential species and
elementary steps needed for a catalytic ionic hydrogena-
tion occurring with heterolytic fission of H2. It is
important to note that even in the case of 6, where the
reaction with the [isopropylideneiminium] reagent
brought about full proton transfer to 6, the acidity of
the formed dihydride [ReH2]+ species could still be in a
suitable range to allow a remaining protolysis equilib-
rium between the hydride and the iminium salt. These
exploratory iminium experiments provided further in-
dication that the given species 4, 5, and 6 could act as
precatalysts for the sought hydrogenations of imines.
3. Ca ta lytic Hyd r ogen a tion of Im in es w ith th e
Com p lexes 4(P h Cl), 5(P h Cl), a n d 6. (1) Low -P r es-
su r e (1-2 ba r ) Ca ta lytic Hyd r ogen a tion of Im in es
by 5 a n d 6 in th e P r esen ce of H2 P u r su ed by NMR.
Hydrogenations of 5(P h Cl) and 6(P h Cl) were explored
applying several simple imines in NMR experiments at
various temperatures starting from 1 bar of H2 at room
temperature to test the catalytic potential and to obtain
further mechanistic details. A mixture of THF/toluene
or PhCl was used as solvent. The results and conditions
are summarized in Table 5.
(2) Catalytic Hydr ogen ation of Im in es by 4(P h Cl)
a n d 5(P h Cl) w ith H2 u n d er P r essu r e. In Table 6 the
conditions for several autoclave hydrogenations using
4(P h Cl) and 5(P h Cl) as catalyst precursors are re-
ported, showing that especially with 5(P h Cl) quite high
yields of the amines can be obtained and also shorter
reaction times are required, even shorter than those in
the NMR tube experiments. It should additionally be
mentioned that the catalyst loading could be reduced.
Solvents of these reactions were not varied greatly
(Table 6), since no drastic effects were seen in explor-
atory reactions. The most effective catalysis was ac-
complished with the catalyst precursor 5(P h Cl) in the
hydrogenation of PhCHdNPh, where at a pressure of
65 bar of H2 and at 35 °C the imine conversion was
almost 100% within just 2 h. Some experiments carried
out with 6 under H2 pressure did not show greatly
improved activities and yields in comparison with those
of Table 5. This led us to assume that this catalyst
system has principal deficiencies (see below), which
could not easily be mended by further tuning efforts.
On the basis of the resulting amounts of the amine
products it is clear that all the hydrogenations of Table
5 were catalytic, differing, however, in the efficiency.
The results for complex 5(P h Cl) were most promising.
As the hydrogenation proceeded, signals of complex 9
became progressively observable; that is, the amine-
coordinated compound was appearing. The latter was
indeed thermally stable for a long period of time at 90
Scheme 4 displays the anticipated mechanism of the
catalytic transformations. As described before, the
crucial uptake of H2 is assumed to occur via formation
of nonclassical H2 complexes, which can appear in
experiments under pressure in substantially higher

3160 Organometallics, Vol. 23, No. 13, 2004
Liu et al.
Ta ble 6. Resu lts of Ca ta lytic Hyd r ogen a tion of Im in es u n d er H2 P r essu r e Usin g 4(P h Cl) a n d 5(P h Cl) a s
Ca ta lysts
cat./imine
(mol %)
temp
(°C)
H2
(bar)
time
(h)
TOF
yield (%)
of amine
cat.
imine
solvent
C6D5Cl
(h-1
)
4(P h Cl)
PhCHdNCH2Ph
0.39
70
90
90
70
90
70
90
75
90
75
90
75
20
65
100
100
55
14
14
100
100
40
63
23
20
22
20
44
41
43
43
43
65
18
29.5
2
1.79
5.50
2
43.9
49.3
100
22
95
26
100
14.3
43.9
27
5(P h Cl)
5(P h Cl)
Me2CdNPri
2.5
1.18
THF/Tol (1:1)
THF/Tol (1:1)
0.85
2.64
1.34
5.54
0.20
2.00
1.34
2.45
9.66
2.96
81.78
5(P h Cl)
5(P h Cl)
5(P h Cl)
0.44
0.51
0.47
C6D5Cl
PhCHdNCH2Ph
PhCHdNCH2Ph
C6D5Cl
C6D5Cl
C6D5Cl
40
100
100
40
35
35
75
5(P h Cl)
5(P h Cl)
PhCHdNPh
PhCHdNPh
0.575
0.587
C6D5Cl
C6D5Cl
100
51.2
96
(CO)3(PMe3)2(PhCl,THF)]+ and [Re(CO)2(PMe3)3(PhCl,-
THF)]+ were synthesized and structurally characterized.
A related [Re(CO)(PMe3)4]+ compound stayed “naked”
as a 16e- species. Testing catalytic hydrogenation reac-
Sch em e 4
tions of different imine substrates with the [Re-
+
(CO)n(PMe3)5-n
]
series of complexes, it could be dem-
onstrated for the first time that the solvent-stabilized
[Re(CO)n(PMe3)5-n(solvent)]+ cations are suitable for
such catalyses. The best catalyst turned out to be the
[Re(CO)2(PMe3)3(solvent)]+ system. The [Re(CO)(PMe3)4]+
species showed low catalytic activity. All these observa-
tions were explained on the basis of an “ionic hydroge-
nation” pathway occurring with heterolytic cleavage of
H2. For optimum catalysis a subtle balance between the
counteracting effects of the strength of back-bonding of
the H2 ligand and its acidity is required, which appar-
ently is satisfied best by the phosphine-trisubstituted
system [Re(CO)2(PMe3)3(solvent)]+.
Exp er im en ta l Section
The preparative work was carried out under nitrogen by
standard Schlenk techniques or in a glovebox. All solvents
were dried over appropriate drying reagents and distilled
under nitrogen before use. Reagent-grade chemicals were used
as purchased from Aldrich or Fluka. [Ph3C][B(3,5-(CF3)2C6H3)4],
B(C6F5)3, isopropylisopropylideneimine, ReH(CO)3(PMe3)2, ReH-
(CO)2(PMe3)3, and cis-ReH(CO)(PMe3)4 were prepared as de-
scribed in the literature. NMR spectra were obtained on a
Varian Gemini-300 spectrometer, measured at 300.23 MHz for
1H NMR and 121.474 MHz for 31P NMR spectroscopy. Chemi-
cal shifts are given in ppm, relative to partially deuterated
solvents peaks (1H) or external 85% H3PO4 aqueous solution
(31P NMR), with downfield shifts considered positive. T1min
measurements were performed on a Bruker WM-500 spec-
trometer, using a standard 180°-τ-90° inversion-recovery
pulse sequence. IR spectra were recorded on a Bio-Rad FTS-
45 spectrometer. Hydrogenations under pressure were per-
formed in a stainless steel autoclave (20 mL total volume) with
concentrations. In the presence of imines the neutral
hydrides and the iminium salts might then be formed
by proton transfer. Subsequently this step could be
followed by imine insertions to generate the amine
complexes of type 8 or 9. In a last step the catalytic cycle
of the “ionic hydrogenation” closes up by replacement
of the amine ligand with H2.
We believe that the irreversible formation of a dihy-
dride compound from the H2 complex reduces catalytic
activity in hydrogenations with heterolytic splitting of
H2, since the proton transfer ability is expected to be
reduced. The high catalytic activity of 5(P h Cl) appar-
ently correlates with the ability to form a relatively
stable and acidic H2 complex as prerequisites for effec-
tive H2 heterolysis. While insertions of the iminium
cations into the Re-H bond to generate amine com-
plexes are anticipated to be “easygoing” for all the
investigated systems, the subsequent and presumably
even rate-determining replacement of the amine by a
solvent molecule and/or a H2 molecule to regenerate the
starting species could lead to retardation of the whole
catalysis depending on the type of amine (or imine).
1
internal stirring. The conversion progress was checked by H
NMR analysis.
Syn th esis of [Re(CO)3(P Me3)2(S)][B(3,5-(CF 3)2C6H3)4]
{S ) P h Cl, 4(P h Cl); S ) THF , 4(THF )}. A solution of 17 mg
(0.04 mmol) of ReH(CO)3(PMe3)2 in 1 mL of PhCl or THF and
a solution of 44.5 mg (0.04 mmol) of [Ph3C][B(3,5-(CF3)2C6H3)4]
in 5 mL of PhCl or THF were mixed at room temperature.
The solution first turned colorless immediately, then dark
yellow after several minutes. The solvent was removed in
vacuo, and the resulting pale yellow residue was washed with
pentane to give off-white solids of 4(P h Cl) or 4(THF ). Crystals
of 4(P h Cl) suitable for the X-ray structure determination were
Con clu sion s
In the series of the cationic complexes [Re(CO)n-
+
(PMe3)5-n
]
the solvent-coordinated compounds [Re-

Carbonyl Rhenium Cations
Organometallics, Vol. 23, No. 13, 2004 3161
recrystallized from a minimum amount of PhCl at -30 °C. 4-
(P h Cl) gave poor results in the elemental analyses presum-
ably due to solvent losses during processing.
MHz, PhCl-d5, 293 K): δ -42.6 (dt, J PP ) 25, 17 Hz, 1P), -47.9
(dt, J PP ) 25, 23 Hz, 1P), -49.2 (dd, J PP ) 23, 17 Hz, 2P).
Anal. Calcd for C45H48ReBF24OP4: C, 39.11; H, 3.50. Found:
C, 38.81; H, 3.39.
4(P h Cl): Yield 42.5 mg (76%). 1H NMR (300.23 MHz, PhCl-
d5, 293 K): δ 1.21 (s, br, Me, 18H). 31P NMR (121.474 MHz,
PhCl-d5, 293 K): δ -33.1 (s, PMe3). 13C NMR (125.783 MHz,
PhCl-d5, 243 K): δ 192.1 (t, J PC ) 8.5 Hz, 2CO), 187.6 (t, J PC
) 6 Hz, 1CO), 18.1 (t, J PC ) 17 Hz, PMe3). IR (Nujol): 2066w,
1967s, 1941s. Anal. Calcd for C47H35ReBClF24O3P2: C, 40.37;
H, 2.52. Found: C, 37.44; H, 6.49.
Syn th esis of [Re(CO)(P Me3)4][BH(C6F 5)3]. A solution of
20 mg (0.0385 mmol) of ReH(CO)(PMe3)4 in 1 mL of PhCl or
THF and a solution of 19.7 mg (0.0385 mmol) of B(C6F5)3 in 5
mL of PhCl or THF were mixed at room temperature. The
solution immediately turned dark yellow. The solvent was
removed in vacuo, and the resulting dark yellow oily residue
was washed with pentane to give a dark yellow oil. Yield: 28
mg (71%). A satisfactory elemental analysis could not be
4(THF ): Yield 31.4 mg (58%). 1H NMR (300.23 MHz, THF-
d8, 293 K): δ 1.30 (s, br, Me, 18H). 31P NMR (121.474 MHz,
THF-d8, 293 K): δ -26.8.0 (s, PMe3). 13C NMR (125.783 MHz,
1
obtained. H NMR (300.23 MHz, PhCl-d5, 293 K): δ 4.25 (q,
br, HB, 1H), 1.07 (t, J PH ) 3 Hz, Me, 18H), 0.95 (d, J PH ) 7.8
THF-d8, 223 K): δ 195.1 (t, J PC ) 9 Hz, 2CO), 193.2 (t, J PC
)
1
Hz, Me, 9H), 0.90 (d, J PH ) 6.9 Hz, Me, 9H). H NMR (300.23
5 Hz, 1CO), 18.5 (t, J PC ) 17 Hz, PMe3). Anal. Calcd for C45H38
-
MHz, THF-d8, 293 K): δ 3.80 (partial overlap with THF, q,
ReBF24O4P2: C, 39.81; H, 2.82. Found: C, 39.37; H, 2.93.
br, HB, 1H), 1.70 (t, J PH ) 3.3 Hz, Me, 18H), 1.59 (d, J PH
)
Syn th esis of [Re(CO)2(P Me3)3(S)][B(3,5-(CF3)2C6H3)4] (S
) P h Cl, 5(P h Cl); S ) THF , 5(THF )). A solution of 16 mg
(0.034 mmol) of ReH(CO)2(PMe3)3 in 1 mL of PhCl or THF and
a solution of 37.6 mg (0.034 mmol) of [Ph3C][B(3,5-(CF3)2C6H3)4]
in 5 mL of PhCl or THF were mixed at room temperature.
The solution first turned colorless and then dark yellow after
several minutes. The solvent was removed in vacuo, and the
resulting pale yellow residue was washed with pentane to give
off-white solids of 5(P h Cl) or 5(THF ). Crystals of 5(P h Cl)
suitable for X-ray structure determination were recrystallized
with a minimum amount of PhCl at -30 °C.
4.5 Hz, Me, 9H), 1.57 (d, J PH ) 3.3 Hz, Me, 9H). 31P NMR
(121.474 MHz, PhCl-d5, 243 K): δ -42.0 (dt, J PP ) 25, 17 Hz,
1P), -48.3 (dt, J PP ) 25, 23 Hz, 1P), -49.3 (dd, J PP ) 23, 17
Hz, 2P). 31P NMR (121.474 MHz, PhCl-d5, 293 K): δ -42.9
(dt, J PP ) 25, 17 Hz, 1P), -48.0 (dt, J PP ) 25, 23 Hz, 1P), -49.0
(dd, J PP ) 23, 17 Hz, 2P).
Rea ction s of 4(P h Cl), 5(P h Cl), a n d 6 w ith 1 ba r of H2:
F or m a tion of Dih yd r ogen a n d Dih yd r id e Com p lexes. A
solution of the cation 4(P h Cl) (12.9 mg, 0.01 mmol), 5(P h Cl)
(13.3 mg, 0.01 mmol), or 6 (13.8 mg, 0.01 mmol) in 0.6 mL of
C6D5Cl was transferred to a Young NMR tube and was frozen
in liquid N2. The N2 atmosphere was substituted by H2. The
[Re(CO)2(P Me3)3(P h Cl)][B(3,5-(CF 3)2C6H3)4] (5): Yield
33.6 mg (68%). 1H NMR (300.23 MHz, PhCl-d5, 293 K): δ 1.17
(t, J PH ) 3.6 Hz, Me, 18H), 1.13 (d, J PH ) 7.5 Hz, Me, 9H).
31P NMR (121.474 MHz, PhCl-d5, 293 K): AB2 system, δ
1
mixture was allowed to warm to -30 °C. The H NMR spectra
were recorded at -30 and -10 °C, respectively.
4(P h Cl)/H2. 1H NMR (500.23 MHz, PhCl-d5, 243 K): δ -
5.70 (s, br, Re(H2), 2H, T1 ) 13 ms). 31P NMR (202.49 MHz,
PhCl-d5, 243 K): δ -44.8 (s, PMe3, 2P).
-34.9, -35.7 (J AB ) 24 Hz, PMe3). Anal. Calcd for C49H44
-
ReBClF24O2P3: C, 40.69; H, 3.07. Found: C, 40.12; H, 2.93.
[Re(CO)2(P Me3)3(THF )][B(3,5-(CF 3)2C6H3)4]: Yield 37.4
5(P h Cl)/H2. 1H NMR (500.23 MHz, PhCl-d5, 243 K): δ -
1
mg (76%). H NMR (300.23 MHz, THF-d8, 293 K): δ 1.30 (t,
J PH ) 3.5 Hz, Me, 18H), 1.25 (d, J PH ) 7.2 Hz, Me, 9H). 31P
NMR (121.474 MHz, THF-d8, 293 K): δ -30.4 (d, J PP ) 25
5.76 (s, br, Re(H2), 2H, T1 ) 12.7 ms), - 5.83 (dd, br, J PH
)
52, 44 Hz, ReH2, 2H, T1 ) 358.8 ms). 31P NMR (202.49 MHz,
PhCl-d5, 243 K): AB2 system, δ -45, -46 (J AB ) 24 Hz, Re-
(H2)PMe3), AB2 system, δ -28.5, -30.8. (J AB ) 24 Hz, ReH2-
PMe3).
Hz, PMe3), -40.2 (t, J PP ) 25 Hz, PMe3). Anal. Calcd for C47H47
ReBF24O3P3: C, 40.15; H, 3.37. Found: C, 40.34; H, 3.15.
-
Syn th esis of [Re(CO)2(P Me3)3(S)][BH(C6F5)3] (S ) P h Cl;
S ) THF ). A solution of 16 mg (0.034 mmol) of ReH(CO)3-
(PMe3)2 in 1 mL of PhCl or THF and a solution of 17.5 mg
(0.034 mmol) of B(C6F5)3 in 5 mL of PhCl or THF were mixed
at room temperature. The solution immediately turned dark
yellow. The solvent was removed in vacuo, and the resulting
dark yellow oily residue was washed with pentane to result
in dark yellow oils. Satisfactory elemental analyses could not
be obtained.
1
6/H2. H NMR (300.23 MHz, PhCl-d5, 243 K): δ - 6.23 (m,
ReH2, 2H, T1 ) 342.14 ms), -7.00 (m, ReH2, 2H, T1 ) 259.2
ms). 31P NMR (121.474 MHz, PhCl-d5, 243 K): δ -21.9 (td,
2
2J PP ) 19.4, 16.5 Hz, PMe3, 1P), -26.2 (td, J PP ) 188.5, 28.4
2
Hz, PMe3, 1P), -36.9 (dd, J PP ) 188.5, 29.9, 48.9, PMe3, 1P),
2
-41.2 (td, 2J PP ) 27.4, 14.0 Hz, PMe3, 2P), -45.2 (dd, J PP
)
2
27 Hz, 22.4 Hz, PMe3, 2P), -51.4 (td, J PP ) 25.4, 15.5 Hz,
PMe3, 2P).
[Re(CO)2(P Me3)3(P h Cl)][BH(C6F 5)3]: Yield 22 mg (58%).
1H NMR (300.23 MHz, PhCl-d5, 293 K): δ 1.25 (t, J PH ) 3.4
Hz, PMe3, 18H), 1.18 (d, J PH ) 7.2 Hz, PMe3, 9H). 31P NMR
(121.474 MHz, PhCl-d5, 293 K): ABX system, δ -35.36(A),
-35.35(B), -36.38(X) (J AB ) 0.4 Hz, J AX ) 23.6 Hz, J BX ) 24.0
Hz).
Rea ction s of ReH(CO)3(P Me3)2, ReH(CO)2(P Me3)3, a n d
ReH(CO)(P Me3)4 w ith [Isop r op ylisop r op ylid en eim in i-
u m ][BAr F ]. A solution of hydride 1 (15 mg, 0.0355 mmol), 2
(16.7 mg, 0.0355 mmol), or 3 (18.4 mg, 0.0355 mmol) in 1 mL
of C6H5Cl was transferred to a screw-cap NMR tube containing
34.2 mg (0.0355 mmol) of [isopropylisopropylideneiminium]-
[BArF]. The solutions turned yellow. The reactions were
monitored by 31P NMR. After 4 h the solvent was removed in
vacuo and the resulting yellow residue was washed with
pentane to give complex 8, 9, or [ReH2(CO)(PMe3)4][BArF] as
off-white solids in quantitative yields.
[Re(CO)2(P Me3)3(THF )][BH(C6F 5)3]: Yield 23 mg (64%).
1H NMR (300.23 MHz, THF-d8/C6D6, 293 K): δ 1.71 (t, J PH
)
3 Hz, PMe3, 18H), 1.15 (d, J PH ) 7 Hz, PMe3, 9H). 31P NMR
(121.474 MHz, THF-d8/C6D6, 293 K): δ -29.8 (d, J PP ) 24 Hz,
PMe3, 1P), -39.8 (t, J PP ) 24 Hz, PMe3, 2P).
[Re(CO)3(P Me3)2(HNiP r 2)][BAr F ] (8). 1H NMR (300.23
MHz, PhCl-d5, 293 K): δ 2.80 (sept, J HH ) 6.6 Hz HC, 2H),
Syn th esis of [Re(CO)(P Me3)4][B(3,5-(CF 3)2C6H3)4] (6).
A solution of 20 mg (0.0385 mmol) of ReH(CO)(PMe3)4 in 1
mL of PhCl or THF and a solution of 42.6 mg (0.0385 mmol)
of [Ph3C][B(3,5-(CF3)2C6H3)4] in 5 mL of PhCl or THF were
mixed at room temperature. The solution immediately turned
colorless, then dark yellow after several minutes. The solvent
was removed in vacuo, and the resulting pale yellow residue
was washed with pentane to give an off-white solid. Yield: 44
mg (82%). Complex 6 can be recrystallized from a minimum
amount of PhCl at -30 °C. 1H NMR (300.23 MHz, PhCl-d5,
293 K): δ 1.18 (t, J PH ) 3 Hz, Me, 18H), 1.05 (d, J PH ) 7.8 Hz,
Me, 9H), 0.98 (d, J PH ) 6.9 Hz, Me, 9H). 31P NMR (121.474
i
1.46 (t, J PH ) 3.3 Hz, Me, 18H), 0.90 (d, J HH ) 6.6 Hz, Pr,
12H). 31P NMR (121.474 MHz, PhCl-d5, 293 K): δ -34.2 (s,
PMe3, 2P). 13C NMR (125.783 MHz, PhCl-d5, 293 K): δ 47.4
(s, HNC(Me)2, 1C), 21.4 (s, Me(CNH), 4Me), 18.4 (t, J PC ) 27
Hz, PMe3, 6Me). Anal. Calcd for C47H45ReBF24NO3P2: C, 40.73;
H, 3.27. Found: C, 40.32; H, 3.38.
[Re(CO)2(P Me3)3(HNiP r 2)][BAr F ] (9). 1H NMR (300.23
MHz, PhCl-d5, 293 K): δ 2.69 (sept, J HH ) 6.6 Hz HC, 2H),
1.42 (t, J PH ) 3.3 Hz, Me, 18H), 1.21 (d, J PH ) 7.5 Hz, Me,
9H), 0.96 (d, J HH ) 6.6 Hz, iPr, 12H). 31P NMR (121.474 MHz,

3162 Organometallics, Vol. 23, No. 13, 2004
Ta ble 7. Cr ysta llogr a h ic Deta ils of 4(P h Cl), 5(P h Cl), 6, a n d 5(THF )'
Liu et al.
4(P h Cl)
5(P h Cl)
6
5(THF )
empirical formula
color
Mr
C
53H40BCl2F24O3P2Re
C49H44BClF24O2P3Re
yellow
1446.21
C45H48BF24OP4Re
brown-yellow
1381.72
C43H63BO4P3Re
yellow
933.85
yellow
1510.70
cryst size (mm)
T (K)
0.72× 0.76 × 0.79
0.12× 0.26 × 0.46
0.24× 0.26 × 0.41
0.07× 0.39 × 0.42
183(2)
193(2)
183(2)
193(2)
λ(Mo KR) (Å)
cryst syst
space group
a (Å)
b (Å)
c (Å)
0.71073
triclinic
P1h
0.71073
monoclinic
C2/c
32.678(2)
13.4212(7)
26.6501(12)
90
99.015(7)
90
11547.1(11)
8
1.664
2.349
5696
0.71073
monoclinic
P2/n
12.4387(10)
12.6647(6)
19.0231(15)
90
109.176(9)
90
2830.5(3)
2
1.621
2.372
1364
0.71073
monoclinic
P21/c
19.4056(14)
9.7996(14)
25.2294(18)
90
112.109(8)
90
4445.0(5)
4
1.395
2.880
1912
12.8032(11)
14.2668(12)
18.0643(14)
111.148(9)
93.531(10)
102.851(10)
2963.6(4)
2
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
Fcalcd (g‚cm-3
)
1.693
2.311
1484
µ (mm-1
F(000)
)
transmn range
θ range
0.2874-0.2398
4.9° < 2θ > 56.88°
29 713
13 723
10 399
0.8542-0.6610
4.6° < 2θ < 55.9°
31 278
9069
6366
0.6680-0.5438
4.72° < 2θ < 56.10°
23 289
6621
4595
0.8239-0.3776
5.34° < 2θ < 60.64°
45 126
12 823
8306
no. of measd reflns
no. of unique reflns
I > 2σ(I) reflns
no. of params
788
1.030
684
1.073
320
1.081
478
1.014
GOF (for F2)
R1[I > 2σ(I)], R1 all data
wR2[I > 2σ(I)], wR2 all data
∆Fmax/min
0.0451, 0.0613
0.1042, 0.1093
1.528, -1.443
0.0527, 0.0736
0.1291, 0.1377
0.967, -0.781
0.0765, 0.1025
0.2215, 0.2314
1.102, -1.730
0.0545, 0.0795
0.1264, 0.1311
3.497, -2.301
R1 ) ∑(Fo - Fc)/∑Fo; I > 2σ(I); wR2 ) {∑w(Fo - Fc2)2/∑w(Fo2)2}1/2
.
a
2
PhCl-d5, 293 K): δ -35.4 (d, J PP ) 27 Hz, PMe3, 2P), -37.8 (t,
J PP ) 27 Hz, PMe3, 1P). 13C NMR (125.783 MHz, PhCl-d5, 293
K): δ 47.3 (s, HNC(Me)2, 1C), 21.1 (s, Me(CNH), 4Me), 19.2
(t, J PC ) 25 Hz, PMe3, 6Me), 17.8 (d, J PC ) 25 Hz, PMe3, 3Me).
stream of an Oxford cryogenic system at the diffractometer.
Measurement temperatures of 183(2), 193(2), 183(2), and 193-
(2) K were used for these structures. An imaging plate detector
system (Stoe IPDS) with graphite-monochromated Mo KR
radiation (λ ) 0.71073 Å) was used for the exposure of 154,
250, 164, and 225 images at constant times of 1.0, 5.0, 4.0,
and 9.0 min per image. The crystal-to-image distances were
set to 58, 60, 60, and 50 mm (θmax ) 28.44°, 27.95°, 28.05°,
and 30.32°). φ-rotation for 4(P h Cl) and φ-oscillation modes
for 5(P h Cl), 6, and 5(THF ) were necessary for the increments
of 1.3°, 0.8°, 1.1°, and 0.8° per exposure in each case. For the
cell parameter refinements 4998 [5(P h Cl)] and 5000 5(THF )62
and 800063 reflections for compounds 4(P h Cl) and 6 were
selected out of the whole limiting spheres with intensities I >
6σ(I). A total of 29 713, 31 278, 23 289, and 45 126 reflections
were collected, of which 13 723, 9069, 6621, and 12 823 were
unique after performing absorption corrections and data
reductions (Rint ) 4.02%, 7.95%, 5.01%, and 14.99%). Nineteen,
16, 16, and 10 indexed crystal faces were used for the
numerical absorption corrections.64
For crystals of 4(P h Cl) including a chlorobenzene solvate
molecule, Patterson and direct methods failed to solve the
structure in space group P1h, but direct methods allowed the
structure solution in space group P1. A center of symmetry
between independent molecules and shifts of 0.112x, 0.140y,
and 0.069z along the unit cell directions allowed final refine-
ment in space group P1h.
The structure of 5(P h Cl) was solved in space group C2/c.
Due to technical reasons we were not able to obtain the full
data set (completeness 62.6%). One of the CF3 groups showed
very large split atomic positions. Fourteen distance restraints
(C-C, C-F, and F-F) were necessary for the isotropic
refinement of this disordered CF3 group using EADP and
PART options in the refinement program SHELXL-97. The
Anal. Calcd for
Found: C, 40.87; H, 3.61.
[ReH2(CO)(P Me3)4][BAr F ]. H NMR (300.23 MHz, PhCl-
d5, 233 K): δ 1.39 (2, J HH ) 9.9 Hz, Me, 9H), 1.21 (d, J PH
C49H54ReBF24NO2P3: C, 41.01; H, 3.79.
1
)
6.3 Hz, Me, 18H), 0.96 (d, J PH ) 7.2 Hz, Me, 9H), -6.38 (m,
ReH2, 2H). 31P NMR (121.474 MHz, PhCl-d5, 233 K): δ -22.8
(dt, J PP ) 21, 15 Hz, PMe3, 1P), -46.1 (dd, J PP ) 27, 21 Hz,
PMe3, 2P), -52.3 (dt, J PP ) 27, 15 Hz, PMe3, 1P). 13C NMR
(125.783 MHz, PhCl-d5, 233 K): δ 26.9 (d, J PC ) 32 Hz, PMe3,
1P), 23.6 (t, J PC ) 19 Hz, PMe3, 2P), 23.1 (d, J PC ) 29 Hz,
PMe3, 1P). Anal. Calcd for C45H50ReBF24OP4: C, 39.06; H, 3.64.
Found: C, 39.51; H, 3.89.
Ca ta lytic Hyd r ogen a tion of Im in es by 5(P h Cl) a n d 6
u n d er 1-2 ba r of H2 Atm osp h er e. Complex 5(P h Cl) or 6
was mixed with imines in 0.6 mL of a mixture of THF-d8/Tol-
d8 or pure C6D5Cl. The solution was transferred to a Young
NMR tube and was frozen. The N2 atmosphere was substituted
by H2. The mixture was warmed to the requested temperature,
and the NMR spectrum was recorded. The conversions were
1
determined by the integration of the imine and the amine H
NMR signals.
Ca ta lytic Hyd r ogen a tion of Im in es w ith 4(P h Cl) a n d
5(P h Cl) u n d er H2 P r essu r e. These reactions were carried
out as following: complex 4 or 5 was dissolved in 1 mL of a
mixture of THF-d8/Tol-d8 or pure C6D5Cl in a steel autoclave
(20 mL), and the imine was added. The reaction vessel was
then pressurized with H2 with pressures according to Table
6. The contents were heated and stirred for given periods of
times. The mixtures were then transferred to NMR tubes, and
1H NMR spectra were recorded. The conversions were deter-
mined by the integration of the imine and the amine signals.
The procedures were repeated until the reactions had stopped.
X-r a y Str u ctu r e Deter m in a tion s. Crystallographic data
for compounds 4(P h Cl), 5(P h Cl), 6, and 5(THF ) are collected
in Table 7. The crystals were embedded in polybutene oil and
mounted on glass fibers, and they were fixed by a cold N2-gas
(62) STOE-IPDS Software package, Version 2.87 5/1998; STOE &
Cie, GmbH: Darmstadt, Germany, 1998.
(63) STOE-IPDS Software package, Version 2.92 5/1999; STOE &
Cie, GmbH: Darmstadt, Germany, 1999.
(64) Coopens, P.; Leiserowitz, L.; Rabinovich, D. Acta Crystallogr.
1965, A18, 1035.

Carbonyl Rhenium Cations
Organometallics, Vol. 23, No. 13, 2004 3163
phenyl group of the chlorobenzene ligand had to be refined
with isotropic displacement parameters.
higher symmetry, and they were helpful to complete the
structure by interpreting the difference electron density
results.
Crystallographic data (excluding structure factors) for 4-
(P h Cl), 5(P h Cl), 6, and 5(THF ) have been deposited with
the Cambridge Crystallographic Data Centre as supplemen-
tary publication nos. CCDC 219386, 219387, 219388, and
219389. Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge CB2 1EZ,
UK, (fax:(+44) 1223-336033; e-mail: deposit@ccdc.cam.ac.uk).
The structure of 6 crystallizes in monoclinic space group P2/
n. In space group P2/n, the 2-fold axes of the molecules coincide
with crystallographic 2-axes. The BArF anion shows now only
one rotational disorder of one (×2) CF3 group; however, every
phosphine group and also the CO group were found on
symmetry-related positions, leading to a 10-coordinated dis-
ordered Re complex. The refinement was performed with
PART and EADP instructions for the disordered trimethyl
phosphine groups. All methyl carbon atoms were isotropically
refined. The disordered ratio was finally 0.51:0.49. The ob-
served disorder for the refined structure 6 in space group P2/n
may be the result of an unresolved merohedric twinning
problem. Four different distance restraints (P-C, C-C) were
applied in the refinement. Compound 5(THF ) crystallized as
THF solvate in extremely thin plates. Low crystal quality may
be the reason for the relatively high Rint value of 15% after
absorption correction and data reduction.
Ack n ow led gm en t. Funding from the Swiss Na-
tional Science Foundation (SNSF) and from the Uni-
versity of Zu¨rich is gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Details of crystal
structure determination for compounds 4(P h Cl), 5(P h Cl), 6,
and 5(THF ). This material is available free of charge via the
Internet at http://pubs.acs.org.
All structures were solved with SHELXS-97.66 The structure
refinements were performed with SHELXL-97.67 Programs
PLATON65 and PLUTON68 were used to check structures for
OM034196Y
(67) Sheldrick, G. M. SHELXL-97, Program for refinement of crystal
structure; University of Go¨ttingen: Germany, 1997.
(68) Spek, A. L. PLUTON, A computer program for the graphical
representation of crystalographic models; University of Utrecht: The
Netherlands, 1991-1997.
(65) Spek, A. L. Acta Crystallogr. 1990, A46, 34.
(66) Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467.