Synthesis of rhenium complexes that contain the [(C6F5NCH2CH2)3N] 3- ligand

Article abstract of DOI:10.1021/om980220+

The reaction between [Et₄N]₂[ReOCl₅] and (C₆F₅NHCH₂CH₂)₃N (H₃[N₃N_F]) in CH₃CN at room temperature in the presence of NEt₃ yielded air stable, emerald green, diamagnetic [(C₆F₅NCH₂CH₂)₂NCH ₂CH₂NHC₆F₅]Re(O)Cl (1). The reaction between 1 and Ta(CH-t-Bu)(THF)₂Br₃ gave paramagnetic [N₃N_F]ReBr (2). An X-ray structure of a sample of 2 showed it to be analogous to that of [N₃N_F]MoCl. Reduction of 2 by methyllithium (or more conveniently by Mg) in the presence of a variety of two-electron ligands gave complexes of the type [N₃N_F]Re(L) (L = H₂, ethylene, propylene, CO, N₂, phosphines, pyridine, tetrahydrothiophene, acetonitrile, or silanes). An X-ray structure of [N₃N_F]Re(ethylene) showed η²-ethylene to be bound in the apical "pocket" with its C-C axis lying in one of the N_ax^- Re-N_eq planes. Protonation of the PMe₃ complex gave an authentic hydrido phosphine complex, {[N₃N_F]Re(H)(PMe₃)}⁺, but protonation of other phosphine complexes gave species in which coupling between H and P is relatively large (～60 Hz) and therefore that are believed to have some {[N₃N_F]Re(η²-HPR_xH _3-x)}⁺ character. An X-ray study of {[N₃N_F]Re(H)-(PHPh₂)}⁺ confirmed that one proton is terminally bound to phosphorus and that the phosphorus is "off-axis" in order to accommodate the "hydride" in the same plane.

Full text of DOI:10.1021/om980220+

Organometallics 1998, 17, 4077-4089
4077
Syn th esis of Rh en iu m Com p lexes Th a t Con ta in th e
[(C6F 5NCH2CH2)3N]3- Liga n d
Steven M. Reid, Brigitte Neuner, Richard R. Schrock,* and William M. Davis
Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
Received March 24, 1998
The reaction between [Et4N]2[ReOCl5] and (C6F5NHCH2CH2)3N (H3[N3NF]) in CH3CN at
room temperature in the presence of NEt3 yielded air stable, emerald green, diamagnetic
[(C6F5NCH2CH2)2NCH2CH2NHC6F5]Re(O)Cl (1). The reaction between 1 and Ta(CH-t-
Bu)(THF)2Br3 gave paramagnetic [N3NF]ReBr (2). An X-ray structure of a sample of 2 showed
it to be analogous to that of [N3NF]MoCl. Reduction of 2 by methyllithium (or more
conveniently by Mg) in the presence of a variety of two-electron ligands gave complexes of
the type [N3NF]Re(L) (L ) H2, ethylene, propylene, CO, N2, phosphines, pyridine, tetrahy-
drothiophene, acetonitrile, or silanes). An X-ray structure of [N3NF]Re(ethylene) showed
η2-ethylene to be bound in the apical “pocket” with its C-C axis lying in one of the Nax-
Re-Neq planes. Protonation of the PMe3 complex gave an authentic hydrido phosphine
complex, {[N3NF]Re(H)(PMe3)}+, but protonation of other phosphine complexes gave species
in which coupling between H and P is relatively large (∼60 Hz) and therefore that are believed
to have some {[N3NF]Re(η2-HPRxH3-x)}+ character. An X-ray study of {[N3NF]Re(H)-
(PHPh2)}+ confirmed that one proton is terminally bound to phosphorus and that the
phosphorus is “off-axis” in order to accommodate the “hydride” in the same plane.
In tr od u ction
trisubstituted silyl group such as SiEt3) or C6F5
([(C6F5NCH2CH2)3N]3- ) [N3NF]3-).1 One attractive
feature of the [N3NF]3- ligand11 (compared to silyl-
substituted ligands) is the stability of the N-C6F5 bond
toward hydrolysis. [N3NF]MLx complexes have now
been prepared that contain Mo or W,11,12 Ti,13 V,14,15 and
Re.16 The Re complexes are relatively important, since
[(Me3SiNCH2CH2)3N]Re analogues so far have not been
prepared and since the only other examples of Re
complexes that contain a tripodal trianionic ligand
appear to be [N(CH2CH2S)3]3- complexes.17-19 We
report here the syntheses of a variety of rhenium
complexes that contain the [N3NF]3- ligand. A small
part of this chemistry has been communicated in a
preliminary fashion.16
A variety of complexes that contain a triamidoamine
ligand ([(RNCH2CH2)3N]3-) and a metal in the 3+ or
higher oxidation state have been prepared.1-12 Tria-
midoamine ligands usually bind to a transition metal
in a tetradentate manner, thereby creating a sterically
protected, 3-fold-symmetric “pocket” in which only three
orbitals are available to bond to additional ligands in
that pocket, two degenerate p orbitals (approximately
xz and dyz) and a σ orbital (approximately dz ). The fact
2
d
that the two frontier p orbitals in a C3 symmetric
triamidoamine complex are strictly degenerate and
probably essentially pure d orbitals creates an environ-
ment that is especially favorable for formation of a
metal-ligand triple bond.
The two most readily accessible [(RNCH2CH2)3N]3-
ligands are those in which R is SiMe3 (or another
Resu lts a n d Discu ssion
An entry into rhenium triamidoamine chemistry was
discovered through the reaction shown in eq 1. We
(1) Schrock, R. R. Acc. Chem. Res. 1997, 30, 9.
(2) Verkade, J . G. Acc. Chem. Res. 1993, 26, 483.
(3) Plass, W.; Verkade, J . G. Inorg. Chem. 1993, 32, 3762.
(4) Schubart, M.; O’Dwyer, L.; Gade, L. H.; Li, W.-S.; McPartlin,
M. Inorg. Chem. 1994, 33, 3893.
(5) Hill, P. L.; Yap, G. A.; Rheingold, A. L.; Maatta, E. A. J . Chem.
Soc., Chem. Commun. 1995, 737.
(6) Mo¨sch-Zanetti, N. C.; Schrock, R. R.; Davis, W. M.; Wanninger,
K.; Seidel, S. W.; O’Donoghue, M. B. J . Am. Chem. Soc. 1997, 119,
11037.
(7) Schrock, R. R.; Seidel, S. W.; Mo¨sch-Zanetti, N. C.; Dobbs, D.
A.; Shih, K.-Y.; Davis, W. M. Organometallics 1997, 16, 5195.
(8) Schrock, R. R. Pure Appl. Chem. 1997, 69, 2197.
(9) Schrock, R. R.; Seidel, S. W.; Mo¨sch-Zanetti, N. C.; Shih, K.-Y.;
O’Donoghue, M. B.; Davis, W. M.; Reiff, W. M. J . Am. Chem. Soc. 1997,
119, 11876.
(10) Dobbs, D. A.; Schrock, R. R.; Davis, W. M. Inorg. Chim. Acta
1997, 263, 171.
(11) Kol, M.; Schrock, R. R.; Kempe, R.; Davis, W. M. J . Am. Chem.
Soc. 1994, 116, 4382.
propose that emerald green diamagnetic 1 is an 18-
electron species that contains one protonated “arm” of
(13) Schrock, R. R.; Cummins, C. C.; Wilhelm, T.; Kol, M.; Lin, S.;
Reid, S.; Davis, W. M. Organometallics 1996, 15, 1470.
(14) Rosenberger, C.; Schrock, R. R.; Davis, W. M. Inorg. Chem.
1997, 36, 123.
(15) Nomura, K.; Schrock, R. R.; Davis, W. M. Inorg. Chem. 1996,
35, 3695.
(12) Seidel, S. W.; Schrock, R. R.; Davis, W. M. Organometallics
1998, 17, 1058.
(16) Neuner, B.; Schrock, R. R. Organometallics 1996, 15, 5.
S0276-7333(98)00220-9 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 08/11/1998

4078 Organometallics, Vol. 17, No. 18, 1998
Reid et al.
the [N3NF]3- ligand that is not coordinated to Re. A
medium-strength, sharp NH stretch is observed in the
IR spectrum at 3419 cm-1, while fluorine and proton
NMR data are more consistent with a structure that
has Cs symmetry. Crystals suitable for a high-quality
X-ray structure could not be obtained. Two “arm-off”
structures have been structurally characterized so far.
14
In {[(C6F5NCH2CH2)2NCH2CH2NHC6F5]V(O)}2 the
two amido ligands occupy equatorial positions in a
pseudo-trigonal bipyramidal arrangement about the
metal, while [(Me3SiNCH2CH2)2NCH2CH2NMe2]Mo-
(NNSiMe3)Me20 is a monomeric diazenido complex
whose structure is analogous to that proposed for 1.
Several examples of TBP complexes that contain a
diamido/donor ligand and in which the amido nitrogen
atoms occupy “equatorial” coordination positions also
have appeared in the literature in the last several
years.21-28 In 1 the oxo ligand would be pseudo-triply
bound to the metal through the dxz and dyz orbitals (if
the N-Re-O axis is the z axis), which is consistent with
the observed RedO stretch at 910 cm-1, leaving the
equatorial amido nitrogen atoms to form one p bond to
2
2
the metal using the dx -y orbital (assuming Re-Cl is
coincident with the x axis). An 18-electron count
thereby could be achieved with the two electrons on the
metal in the dxy orbital. In this circumstance no ligands
compete with each other for metal orbitals. An 18-
electron pseudo-octahedral species is considered to be
a less likely possibility, since the oxo ligand cannot be
pseudo-triply bound to the metal in that circumstance.
If the amine donor were bound to the Re on the NMR
time scale, NMR spectra also should be considerably
more complex.
F igu r e 1. Plots of øM (top) and µeff (bottom) vs T for [N3NF]-
ReBr.
in 1 with Br2 on Ta, loss of HCl from Re to give “[N3NF]-
ReBr2”, and then reduction of “[N3NF]ReBr2” to [N3NF]-
ReBr. The final form of the tantalum and the mode of
reduction of Re(V) to Re(IV) could not be determined.
We have not been able to obtain any single completely
satisfactory set of analyses for C, H, and N in six
attempts, perhaps in part because of the presence of a
somewhat variable amount (∼0.5 equiv) of dichlo-
romethane (as confirmed by NMR) and the possibility
that some [N3NF]ReCl is present in some samples. (See
the discussion of the X-ray study below and in the
Experimental Section.) The average of six C, H, and N
analyses and two Br and Cl analyses is satisfactory,
however (see the Experimental Section), for a compound
with the composition C24H12N4BrF15Re(CH2Cl2)0.5. Sev-
eral pieces of data are consistent with the proposed
formulation. First, complex 2 should have a low-spin
d3 configuration, since the three electrons reside in
degenerate dxz/dyz orbitals (taking the N-Re-Br axis
as the z axis). A plot of the molar magnetic susceptibil-
ity as a function of temperature shows that 2 does
behave as a Curie paramagnet with µ ) 1.65(1) µB
between 5 and 300 K and that the effective magnetic
moment remains constant in this temperature range
(Figure 1), all consistent with a single unpaired spin in
a d3 complex with degenerate dxz and dyz orbitals. We
originally proposed (on the basis of a single C, H, and
N analysis and observation of a strong peak for {[N3NF]-
ReBr}+ in the FAB mass spectrum, and in the absence
of bromide and chloride analyses and SQUID data) that
2 was {[N3NF]ReBr}Br.16 Compound 2 can be observed
in mixtures from the reaction between ReBr4(tetra-
hydrothiophene)2, H3[N3NF], and NEt3, although we
An attempt to exchange the oxo ligand in 1 with the
neopentylidene ligand in Ta(CH-t-Bu)(THF)2Br329 gave
an olive-green paramagnetic species that we propose is
[N3NF]ReBr (2) in 60-85% yield (eq 2). On paper this
reaction consists of a formal exchange of the oxo ligand
(17) Glaser, M.; Spies, H.; Lu¨gger, T.; Hahn, F. E. J . Organomet.
Chem. 1995, 503, C32.
(18) Spies, H.; Glaser, M.; Pietzsch, H.-J .; Hahn, F. E.; Kintzel, O.;
Lu¨gger, T. Angew. Chem., Int. Ed. Engl. 1994, 33, 1354.
(19) Spies, H.; Glaser, M.; Pietzsch, H.-J .; Hahn, F. E. Inorg. Chim.
Acta 1995, 240, 465.
(20) O’Donoghue, M. B.; Davis, W. M.; Schrock, R. R. Inorg. Chem.,
in press.
(21) Horton, A. D.; de With, J .; van der Linden, A. J .; van de Weg,
H. Organometallics 1996, 15, 2672.
(22) Cloke, F. G. N.; Hitchcock, P. B.; Love, J . B. J . Chem. Soc.,
Dalton Trans. 1995, 25.
(23) Clark, H. C. S.; Cloke, F. G. N.; Hitchcock, P. B.; Love, J . B.;
Wainwright, A. P. J . Organomet. Chem. 1995, 501, 333.
(24) Cloke, F. G. N.; Geldbach, T. J .; Hitchcock, P. B.; Love, J . B. J .
Organomet. Chem. 1996, 506, 343.
(25) Baumann, R.; Davis, W. M.; Schrock, R. R. J . Am. Chem. Soc.
1997, 119, 3830.
(26) Schrock, R. R.; Lee, J .; Liang, L.-C.; Davis, W. M. Inorg. Chim.
Acta 1997, 270, 353.
(27) Blake, A. J .; Collier, P. E.; Gade, L. H.; McPartlin, M.;
Mountford, P.; Schubart, M.; Scowen, I. J . Chem. Commun. 1997, 1555.
(28) Friedrich, S.; Schubart, M.; Gade, L. H.; Scowen, I. J .; Edwards,
A. J .; McPartlin, M. Chem. Ber.-Rec. 1997, 130, 1751.

Synthesis of Rhenium Complexes
Organometallics, Vol. 17, No. 18, 1998 4079
Ta ble 1. Cr ysta llogr a p h ic Da ta , Collection P a r a m eter s, a n d Refin em en t P a r a m eter s for [N3NF ]ReBr (2),
[N3NF ]Re(C2H4) (3a ), a n d {[N3NF ]Re(H)(P HP h 2)}BAr 4 (4gH+)a
2
3a
4gH+
empirical formula
formula weight
C24H12Br0.6Cl0.4F15N6Re
917.72
C26H16N4F15Re
855.63
C68H36N4BF39PRe
1877.99
crystal dimensions (mm)
crystal system
a (Å)
0.19 × 0.14 × 0.08
0.21 × 0.18 × 0.13
0.20 × 0.15 × 0.30
orthorhombic
triclinic
monoclinic
15.3411(1)
11.151(2)
21.0214(1)
b (Å)
18.9146(13)
11.421(2)
17.0496(2)
c (Å)
11.5416(8)
11.937(2)
22.7095(3)
R (deg)
â (deg)
γ (deg)
V (Å3)
90
90
90
105.622(2)
107.363(2)
93.121(2)
90
116.759(1)
90
7267.5(1)
3349.0(4)
1382.2(5)
space group
Z
D
P212121
4
1.820
P1h
P21/c
4
1.716
2
calc (Mg/m3)
2.056
F000
1743
820
3672
temperature (K)
183
293(2)
188(2)
θ range for data collection (deg)
no. of independent reflections
no. of data/restraints/parameters
final R indices [I > 2σ(I)]
R indices (all data)
GOF
1.71-23.41
1.87-23.27
1.56-22.96
4703
3860
9940
4695/0/404
3853/0/415
R1 ) 0.0471, wR2 ) 0.1180
R1 ) 0.0489, wR2 ) 0.1281
1.079
9933/171/1042
R1 ) 0.0632, wR2 ) 0.1395
R1 ) 0.0779, wR2 ) 0.1493
1.180
R1 ) 0.0408, wR2 ) 0.1117
R1 ) 0.0439, wR2 ) 0.1341
1.083
extinction coefficient
0.0014(2)
0.640 and -0.928
(not applied)
2.459 and -2.675
0.00029(6)
1.527 and -1.475
largest diff peak and hole (e Å-3
)
a
All structures were solved on a Siemens SMART/CCD diffractometer using 0.710 73 Å Mo KR radiation and ω scans. Solutions were
refined using a full-matrix least-squares on F2.
Ta ble 2. Selected In tr a m olecu la r Dista n ces (Å)
a n d An gles (d eg) for th e Non -Hyd r ogen Atom s in
[N3NF ]MoCl, [N3NF ]ReX (2), [N3NF ]Re(C2H4) (3b),
a n d {[N3NF ]Re(H)(P HP h 2)}+ (4gH+)
[N3NF]-
MoCla
2
3b
4gH+
M-N(1)
1.962(3) 1.940(8) 1.930(7)
1.957(3) 1.913(8) 1.965(7)
1.964(3) 1.917(8) 1.940(7)
2.182(3) 2.173(8) 2.178(8)
119.0(1) 117.1(3) 116.9(3)
117.4(1) 116.0(3) 118.3(3)
115.9(1) 120.2(4) 117.1(3)
127.8(3) 125.3(6) 129.7(6)
125.9(3) 118.8(6) 130.8(6)
124.6(2) 124.9(7) 130.4(6)
1.949(7)
1.991(7)
1.998(7)
2.198(6)
114.5(3)
121.0(3)
115.4(3)
129.5(5)
133.2(5)
129.0(5)
175
M-N(2)
M-N(3)
M-N(4)
N(1)-M(1)-N(2)
N(2)-M(1)-N(3)
N(1)-M(1)-N(3)
M-N(1)-C(11)
M-N(2)-C(21)
M-N(3)-C(31)
N(4)-M-N(1)-Cb 177
N(4)-M-N(2)-Cb 168
N(4)-M-N(3)-Cb 172
173
174
179
173
172
170
F igu r e 2. Side view (ORTEP) of the structure of [N3NF]-
ReX, where X ) 0.60Br/0.40Cl.
179
161
2.410(2)
(P)
M to ligand
2.367(1) 2.442(2) 2.182(8)
(see the Experimental Section). However, the axial
ligand could not be refined satisfactorily as either Br
or Cl, and some disorder was observed in the ligand
backbone. The backbone disorder could be modeled
readily. However, only refinement of the axial “halide”
as 0.60 Br and 0.40 Cl led to a structure of relatively
high quality. We surmise that the particular crystal
chosen for the X-ray study happened to be a mixture of
∼60% [N3NF]ReBr and ∼40% [N3NF]ReCl, but that this
crystal is not representative of the bulk material.
The structure of [N3NF]ReX (where X ) 0.60 Br and
0.40 Cl; Figure 2) is analogous to that of [N3NF]MoCl,11
as shown by the bond distances and angles listed in
Table 2. The Re-Neq distances in [N3NF]ReX appear
to be slightly shorter by up to 0.02 Å than the Mo-Neq
distances in [N3NF]MoCl, but the M-N(4) distance and
the Neq-Mo-Neq and M-N-C angles in the two com-
pounds are essentially identical. In each compound the
C6F5 rings are turned so as to form a bowl around the
apical halide atom, and the Nax-Re-Neq-C dihedral
angles are all between 168° and 179°. The Re-X
distance is of course longer than an expected Re-Cl
(Cl)
(X)
(C(7))
2.147(11)
1.406(14)
Re-C(8)
C(7)-C(8)
N(4)-Re-P
P-H(1)
160.2(2)
1.39(7)
93.38(6)
103.3(4)
Re-P-H(1)
C(11)-P-C(21)
a
See ref 11. Only one of two essentially identical molecules in
b
the unit cell. Taken from a Chem 3D model.
have not yet been able to synthesize it in pure form and
high yield in this manner.
The most convincing evidence for the identity of 2 is
an X-ray structural determination (Tables 1 and 2;
Figure 2), although this structure was not completed
without complications. In the bulk sample on the
average approximately one Br and one Cl are present,
with the majority of the Cl coming from ∼0.5 equiv of
dichloromethane; however, it should be noted that
analyses tend to be slightly high in Cl and low in Br
(see Experimental Section). The dichloromethane (dis-
ordered) was found in the crystal chosen for the X-ray
study, and in the process was removed using SQUEEZE

4080 Organometallics, Vol. 17, No. 18, 1998
Reid et al.
distance and shorter than an expected Re-Br distance,
given the method of modeling the apical halide ligand.
Dih yd r ogen , Din itr ogen , CO, a n d Olefin Com -
p lexes. Attempts to alkylate 2 with methyllithium
revealed that it is readily reduced to Re(III) and that
in the presence of several two-electron donor ligands
complexes with the formula [N3NF]Re(L) are produced.16
We have since found that 2 is reduced smoothly by Mg
powder and that MgBr2 is easily removed as insoluble
[MgBr2(dioxane)]x. Reduction of 2 by Mg is also more
straightforward and is less likely to be susceptible to
side reactions. Therefore reduction of 2 by Mg was the
method of choice for preparing lower oxidation state Re
species, although yields were not in any case necessarily
greater than when methyllithium was employed as a
reducing agent.
with H2 in toluene-d8 and toluene, respectively. We
have seen no evidence that deuterium scrambles into
the backbone of the [N3NF]3- ligand.
Loss of H2 to give “Re[N3NF]” alone would not explain
the H/D exchange results above. We propose that one
H in 3a migrates to one of the amido nitrogen atoms to
give a monohydride complex (3a ′, eq 3). Both 3a and
Reduction of 2 under dihydrogen gave [N3NF]ReH2
(3a ). A sharp resonance for the “hydrides” in 3a was
found at -0.89 ppm in the proton NMR spectrum in
CD2Cl2, but no peak could be located in the IR spectrum.
A T1 measurement at 21 °C yielded a value of 138 ms,
a possibly nonminimal value that would be more con-
sistent with 3a being a classical dihydride complex.30-32
However, [N3NF]ReHD was prepared similarly and
shown to have an HD coupling constant of 17 Hz, which
indicates that 3a has some characteristics of a dihydro-
gen complex. (Typically J HD ) 30 ( 4 Hz in dihydrogen
complexes.30-32) There are examples in the literature
where neither description (a dihydrogen complex or a
dihydride) is totally satisfactory,33 and we suggest that
is the case here, at least on the basis of the data
available so far. For purposes of discussion we choose
to view 3a as an 18e dihydrogen complex of Re(III) in
which two amido-Re π bonds can form1 and in which
the two orthogonal filled dxz and dyz orbitals contain the
four d electrons. Since the dxz and dyz orbitals are
equivalent in C3v symmetry, back-bonding to the dihy-
drogen ligand does not “lock” the dihydrogen ligand in
any given orientation.30,31,34,35
Approximately 2 equiv of D2 was added to an NMR
tube containing [N3NF]ReH2 in toluene-d8. At room
temperature, the 1H NMR spectrum showed only [N3NF]-
ReH2, but after heating the sealed sample to 120 °C
overnight the H2 resonance at -0.58 ppm was replaced
by a triplet at -0.63 ppm (J HD ) 17 Hz) arising from
[N3NF]ReHD. The same experiment was repeated in
toluene and monitored by 2H NMR, whereupon a
doublet at -0.51 ppm (J HD ) 17 Hz) and a singlet at
-0.60 ppm in approximately a 1:1 ratio were observed
for [N3NF]ReHD and [N3NF]ReD2, respectively. We
ruled out the possibility that solvent was playing a role
in H/D exchange, since no H/D exchange occurred when
[N3NF]ReH2 was heated to 100 °C for 16 h in degassed
toluene-d8. Similar behavior was observed in 1H and
2H NMR experiments in which [N3NF]ReD2 was treated
3a ′ would be 16e Re(III) complexes since the two
equatorial amido nitrogen atoms can form only one π
bond to the metal. Further reaction of 3a ′ with D2
would yield an 18-electron complex, or if the amine
donor dissociates, the 16-electron Re(V) complex 3a ′′.
Loss of HD from the 18-electron or 16-electron Re(V)
complex and reformation of 3a ′-d1 and then 3a would
complete one H/D exchange cycle. Complexes that
contain Re(III), Re(IV), or Re(V) are a central theme in
this paper, and therefore oxidation of Re(III) to Re(V)
is a reasonable step to propose in the exchange process.
An intermediate that contains “one arm off” is also
feasible in view of the proposed structure of 1 and the
known structure of {[(C6F5NCH2CH2)2NCH2CH2-
NHC6F5]V(O)}2.14 Finally, we have recently structur-
ally characterized a d2 molybdenum species related to
3a ′,
{[(Me3SiNCH2CH 2)2NCH 2CH 2NMe2]Mo(Nd
NSiMe3)}+, as well as the product of addition of MeMgCl
to it, trigonal bipyramidal [(Me3SiNCH2CH2)2NCH2CH2-
NMe2]Mo(NdNSiMe3)(Me),
in
which
the
di-
methylamine donor is not bound to the metal.20 There-
fore, both 3a ′ and 3a ′′ are plausible intermediates in
the proposed mechanism of the H/D exchange reaction
shown in eq 3. The possibility that D2 attacks 3a to
give a [N3NF]Re(H2)(D2) intermediate in which H and
D scramble seems less likely to us at this stage than
the mechanism proposed in eq 3.
[N3NF]Re(C2H4) (3b) was prepared by reduction of 2
in THF with magnesium under an atmosphere of
ethylene. A proton NMR spectrum of 3b in THF-d8
revealed a resonance for coordinated ethylene at 2.61
ppm. The proton NMR spectrum of an analogous
compound prepared from doubly 13C labeled ethylene
showed the expected pattern for an AA′A′′A′′′XX′ sys-
tem36 with a J H
coupling of ∼152 Hz. A non-first-
(29) Rupprecht, G. A.; Messerle, L. W.; Fellmann, J . D.; Schrock,
R. R. J . Am. Chem. Soc. 1980, 102, 6236.
(30) J essop, P. G.; Morris, R. H. Coord. Chem. Rev. 1992, 121, 155-
284.
(31) Crabtree, R. H.; Hamilton, D. G. Adv. Organomet. Chem. 1988,
28, 299.
(32) Heinekey, D. M.; Oldham, W. M., J r. Chem. Rev. 1993, 93, 913.
(33) Earl, K. A.; J ia, G.; Maltby, P. A.; Morris, R. H. J . Am. Chem.
Soc. 1991, 113, 3027.
(34) Kubas, G. J . Acc. Chem. Res. 1988, 21, 120.
(35) Eckert, J .; Kubas, G. J .; White, R. P. Inorg. Chem. 1992, 31,
1550-1551.
13C
order multiplet was observed at 54.7 ppm in the 13C
NMR spectrum of [N3NF]Re(13C2H4) in CD2Cl2, and
upon decoupling 1H, a singlet was observed with no
discernible coupling to fluorine. The only significant
1
change in the H NMR spectrum upon cooling an NMR
sample of [N3NF]Re(C2H4) to -90° was resolution of four
(36) Lynden-Bell, R. M. Mol. Phys. 1963, 6, 601.

Synthesis of Rhenium Complexes
Organometallics, Vol. 17, No. 18, 1998 4081
1
Figure 4. H NMRspectrum (CD2Cl2,-90°C)of[N3NF]Re(CH3-
CHdCH2).
The C6F5 groups are oriented in a manner that creates
a bowl-like trigonal cavity in 2 and 3b. In this “bowl-
like” configuration the Fortho‚‚‚Fortho distances are on the
order of 3.1 Å (3.12, 3.08, and 3.11 Å in 3b). Therefore
the trigonal arrangement is maintained to a significant
degree by steric interaction between C6F5 rings. In
contrast, the C6F5 rings in [N3NF]V lie approximately
in the V-Neq-Cipso plane,14 and the short V‚‚‚Fortho bond
distances (2.652 Å) suggest a weak interaction of the
ortho fluorines with the metal. Therefore one might
expect some stabilization of “[N3NF]Re” in a similar
manner, should it form, and conversely a considerable
degree of steric interaction between a C6F5 ring that
approaches an orientation in which it would lie ap-
proximately in the Re-Neq-Cipso plane and any ligand
in the trigonal coordination pocket.
F igu r e 3. (a) Side view (ORTEP) of the structure of [N3NF]-
Re(C2H4). (b) Top view of the structure of [N3NF]Re(C2H4).
A propylene complex, [N3NF]Re(CH3CHdCH2) (3c),
could be prepared in 65% yield as sparingly soluble dark
types of backbone methylene resonances ascribed to the
methylene [N3NF]3- protons as a consequence of the
[N3NF]3- ligand becoming locked in a conformation with
true C3 symmetry. The dxz and dyz orbitals are still
degenerate in the C3 point group, so there is no
“electronic” barrier that would prevent free rotation of
the ethylene ligand. There is no evidence that ethylene
rotation slows on the NMR time scale for steric reasons,
i.e., that a complex with no symmetry is produced at
-90 °C.
green prisms (eq 4).
A
1H NMR spectrum at room
temperature reveals that 3c is 3-fold symmetric and
that the four resonances for a rapidly rotating propylene
ligand are sharp and well-resolved. However, upon
lowering the temperature of the sample, two sets of
propylene resonances are obtained along with two sets
of overlapping resonances characteristic of a C3-sym-
metric ligand backbone (Figure 4). We assign the two
sets of propylene resonances to two diastereomers of
approximately equal energy formed when propylene
binds to the face of the chiral (C3) complex that is formed
at low temperature. In each diastereomer the propylene
is freely rotating on the NMR time scale, since only four
ligand backbone resonances and one type of pentafluo-
rophenyl ring are observed in each diastereomer.
Reduction of 2 under 14N2 or 15N2 yielded [N3NF]Re-
(N2) (3d ) or [N3NF]Re(15N2) (3d -15N2), respectively. IR
The structure of [N3NF]Re(C2H4) was elucidated in an
X-ray study (Tables 1 and 2; Figure 3). The ethylene
ligand lies approximately in the N(4)-Re-N(2) plane.
The C(7)-C(8) bond length (1.406(14) Å) is typical of a
bound ethylene, as are the Re-C(ethylene) bond lengths
(2.182(8) Å and 2.147(1) Å). The Re atom is displaced
above the equatorial nitrogen plane by 0.314 Å. Re-
Neq distances vary from 1.930 to 1.965 Å, Re-Neq-C
angles are all approximately 130°, and the Re-N(4)
distance is 2.18 Å; all are typical for [N3NF]3- complexes
of a second- or third-row metal1 and similar to the
analogous distances and angles in 2. The Re-Neq-Cipso
bond angles are consistently a few degrees larger than
they are in [N3NF]ReX, presumably as a consequence
of greater steric interaction between the C6F5 groups
and ethylene versus a halide. For comparison, it should
be noted that the acetylene ligand in [(Me3SiNCH2-
CH2)3N]Ta(HCtCH) lies in one of the Nax-Ta-Neq
planes with Ta-C ) 2.09(1) and 2.10(1) Å with one end
pointing directly toward one of the trimethylsilyl sub-
stituents on the triamidoamine ligand,37 while the
cyclopentylidene and the hydride ligands in [(Me3-
SiNCH2CH2)3N]W(H)(C5H8) lie in a Nax-W-Neq plane.7
spectra of 3d (νNN ) 2004 cm-1) and 3d -15N2 (ν
)
15N15
N
1935 cm-1) are consistent with an end-on mode of
dinitrogen binding, as is the 15N NMR spectrum of 3d -
15N2, which reveals two doublet resonances at 281.3 and
332.0 ppm with J NN ) 5.5 Hz. Exposure of 3d -15N2 to
1
14N2 (22 °C, 1 atm) does not lead to exchange of 15N2 for
14N2 over a period of 24 h. [N3NF]Re(N2) (d4) is related
to [N3NF]Mo(N2) (d3), which is believed to be an inter-
mediate in reactions involving dinitrogen in the [N3NF]Mo
system,11 but which has not yet been observed. How-
(37) Freundlich, J . S.; Schrock, R. R.; Davis, W. M. Organometallics
1996, 15, 2777.

4082 Organometallics, Vol. 17, No. 18, 1998
Reid et al.
Ta ble 3. P r op er ties of P h osp h in e Com p lexes
complexes known to date in the general category of
triamidoamine complexes.1 The stability of the PH3
complex might be attributed to protection against bi-
molecular decomposition reactions. In 4a and 4g the
value for J HP is relatively large (305 and 319 Hz,
respectively), as expected by comparison to J HP values
in uncomplexed phosphines (180-230 Hz) and phos-
phonium salts (250-730 Hz).41
δ P
θ
δ H
J HP
(Hz)
complex
color
(ppm) (deg)a (ppm)
[N3NF]Re(PH3) (4a )
[N3NF]Re(PMe3) (4b)
[N3NF]Re(PEt3) (4c)
[N3NF]Re[P(i-Pr)3] (4d ) green
[N3NF]Re(PMe2Ph) (4e) red-orange -30.91 122
red
red
red
-138.18 87
305
-45.16 118
-24.92 132
-11.56 160
[N3NF]Re(PMePh2) (4f) red
[N3NF]Re(PHPh2) (4g) orange
{[N3NF]Re(PMe3)(H)}+ green
-15.19 136
1.24 128
-14.88
319
7
Several other complexes that contain σ donor ligands
could be prepared readily, as shown in eq 6. All are
0.26
(4bH+
)
{[N3NF]Re(PEt3)(H)}+ amber
(4cH+
{[N3NF]Re(PMePh2)-
(H)}+ (4fH+
{[N3NF]Re(PHPh2)(H)}+ coral
(4gH+
23.02
23.74
38.11
3.76 64
)
red-brown
4.72 62
)
3.36 54, 375b
)
a
b
See ref 40. J HH ) 15 Hz.
ever, [(Me3SiNCH2CH2)3N]No(N2) (νNN ) 1934 cm-1
)
has been isolated and structurally characterized.20,38
1
Reduction of 2 under CO yields [N3NF]Re(CO) (3e).
The IR spectrum of 3e has a CO band at 1875 cm-1 that
shifts to 1830 cm-1 in [N3NF]Re(13CO). The value for
νCO suggests that the amount of back-bonding is quali-
tatively significantly less that in [N3N]W(H)(CO), for
diamagnetic and 3-fold symmetric on the H NMR time
scale. The pyridine complex was isolated in 48% yield
as dark burgundy needles, the tetrahydrothiophene
complex in 68% yield as brown needles, and the aceto-
nitrile complex in 61% yield as olive-green prisms. The
1H NMR spectrum of 6 in CD2Cl2 showed only one broad
singlet for the THT ligand centered at 1.23 ppm, which
we ascribe to an accidental overlap of the THT R and â
and exo and endo methylene protons; the 13C{1H} NMR
spectrum shows clearly two sharp resonances for the
two types of methylene carbons in the THT ligand. The
IR spectrum of 7 in Nujol shows the CN stretch at 2152
cm-1. Reduction of 2 in THF in the absence of any other
potential ligand resulted only in decomposition; that is,
no [N3NF]Re(THF) complex could be observed. It should
be noted that [N3NF]V(THF) and [N3NF]V(CH3CN) are
known and have been structurally characterized.15
Sila n e Com p lexes. The synthesis of what we pro-
pose is [N3NF]Re(η2-H2) led us to speculate that alkyl-
silane complexes of the general form [N3NF]Re(η2-
alkylsilane) might be stable. Two examples could be
prepared by reduction of 2 in the presence of excess
alkylsilane, as shown in eq 7, and isolated readily as
orange-red crystals. Proton NMR spectra in THF-d8
example, where νCO ) 1766 cm-1 10
The more “normal”
.
value for νCO in 3e is perhaps in part a reflection of the
electron-withdrawing ability of the C6F5 groups relative
to the SiMe3 groups and the consequent reduced back-
bonding or reducing ability of the metal. A d3 tungsten
analogue, [N3NF]W(CO), has been prepared in which νCO
) 1846 cm-1 39
.
σ Don or Com p lexes. Several phosphine complexes
of the general formula [N3NF]Re(PR1R2R3) could be
prepared by treating 2 with excess Mg in the presence
of 2-4 equiv of PR1R2R3 in degassed THF (eq 5). 31P-
{1H} NMR chemical shifts for the complexes and cone
angles40 for the free phosphines are summarized in
Table 3. Mixtures containing a significant quantity of
[N3NF]Re(N2) were generated if these reactions were
performed under an atmosphere of dinitrogen. Addition
of 1,4-dioxane and removal of [MgBr2(dioxane)]x allowed
the diamagnetic phosphine complexes to be isolated as
moderately air-stable orange to red cubes. These spe-
cies are soluble in dichloromethane, THF, and toluene
and exhibit 3-fold symmetry on the NMR time scale in
solution. An isolable PPh3 complex could not be pre-
pared under these conditions, we assume as a conse-
quence of the relatively poor basicity of PPh3 in combi-
nation with its large steric demands. The largest
phosphine that can be accommodated is P(i-Pr)3 (θ )
160°); its steric bulk presumably contributes to its
lability (see below). These are the only phosphine
showed that both have 3-fold symmetry on the NMR
time scale. The resonance for what is assigned to the
η2-Si-H proton in 8a is a singlet at -5.85 ppm with a
value for J SiH (44 Hz) that is in the upper range for η2-
silane complexes.42,43 In contrast, the resonance for the
terminal Si-H proton in 8a is found at 5.83 ppm with
J SiH ) 186 Hz. The methylene resonances in the Et2-
SiH2 ligand are found as two multiplets centered at
-0.01 and -0.74 ppm, consistent with their diaste-
reotopic nature, while the equivalent methyl groups give
(38) O’Donoghue, M. B.; Zanetti, N. C.; Schrock, R. R.; Davis, W.
M. J . Am. Chem. Soc. 1997, 119, 2753.
(39) Seidel, S. W.; Davis, W. M. Unpublished results.
(40) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(41) Brazier, J . F.; Houalla, D.; Loenig, M.; Wolf, R. Top. Phosphorus
Chem. 1976, 8, 99.
(42) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789.
(43) Schubert, U. Adv. Organomet. Chem. 1990, 30, 151.

Synthesis of Rhenium Complexes
Organometallics, Vol. 17, No. 18, 1998 4083
1
rise to one triplet at 0.32 ppm. The H NMR spectrum
of 8b in CD3CN showed that it is similar to 8a in that
the η2-Si-H resonance is found at -3.86 ppm with J SiH
) 38 Hz, while the resonance for the two terminal Si-H
protons is found at 5.00 ppm with J SiH ) 183 Hz.
NMR time scale at 25 °C in CD2Cl2; the hydride
resonance is observed as a doublet at 0.26 ppm (J PH
)
7 Hz; Table 3). The 31P{1H} NMR spectrum indicates
that the phosphorus resonance is shifted downfield from
that in 4b to -14.88 ppm, consistent with the cationic
formulation for 4bH+. The protonation reaction is
reversible; 4bH+ reacts with excess Et3N over a period
of several days at 25 °C to afford 4b quantitatively
(according to 19F NMR).
Transition metal complexes that contain a silane
ligand bound in an η2 fashion are becoming more
numerous.31,42-46 The most reliable indicator for dis-
tinguishing an η2-silane ligand from a silyl hydride is
J
SiH, which for an η2-silane ligand is usually in the range
15-60 Hz and for a silyl hydride complex is less than
15 Hz (if observed at all). In 8a and 8b J SiH ) 44 and
38 Hz, respectively, thus clearly establishing these
complexes as η2-silane complexes. Rhenium silane
complexes of the type (R2SiH2)Re2(CO)8 (R ) Ph or Me)47
appear to have been the first reported examples of Re-
(η2-silane) complexes, and others are now known in the
literature.48-55
Addition of HBAr4 to 4c, 4f, or 4g gave complexes that
were slightly different from 4bH+. The “hydride”
resonance was observed between ∼3 and ∼5 ppm and
was found to be coupled to phosphorus to the extent of
54-64 Hz. These data suggest that the “hydride” ligand
is coupled more strongly to phosphorus than in 4bH+,
but not as strongly as a proton bound to phosphorus
directly (e.g., 305 Hz in 4a and 319 Hz in 4g). The NMR
spectra of the product formed upon protonation of 4g
with HBAr4 were especially revealing. The 1H NMR
spectrum of 4gH+ in CD2Cl2 revealed a double doublet
at 6.97 ppm with J PH ) 375 Hz and J HH ) 15 Hz,
consistent with a phosphorus-bound proton. Another
double doublet at 3.36 ppm with J PH ) 54 Hz was
assigned to the Re “hydride”. Therefore we propose that
phosphine complexes are protonated to give a classical
hydride with a small coupling to P when the phosphine
is PMe3, but a species in which the proton is more
strongly interacting with the phosphorus in the other
examples described here (e.g., eq 9). The “hydride” (Hb)
Liga n d Su bstitu tion Rea ction s. When dichlo-
romethane solutions of various phosphine complexes
(except 4d ) were treated with 3-5 equiv of other
phosphines, THT, or MeCN, and monitored by 19F NMR,
no bound phosphine was displaced by another ligand
at 22 °C over a period of 3 days. Under identical
conditions, the silane ligands in 8a and 8b were not
replaced with H3SiPh and H2SiEt2, respectively. [N3NF]-
Re(N2) also did not react with MeCN, PMe3, or C2H4
under similar conditions. However, experiments in
which solutions of 4d in degassed toluene (28.9 mM,
85-120 °C, 16 h) were treated with other ligands (3
equiv) showed that P(i-Pr)3 is readily replaced by MeCN,
N2, PMe3, and C2H4. Likewise, [N3NF]ReH2 (34.4 mM
in CH2Cl2) reacted with PMe3 (3 equiv) to give 4b after
3 days at room temperature, while [N3NF]Re(propylene)
was converted to 4b in about 10% yield under the same
conditions. More detailed studies will be required in
order to determine whether the exchange reactions are
associative or dissociative. However, the reactivity of
4d seems likely to be ascribable to the lability of the
P(i-Pr)3 ligand (for steric reasons) and formation of
intermediate “[N3NF]Re”.
P r oton a tion Rea ction s. When red [N3NF]Re(PMe3)
is treated with 1 equiv of HBAr4 in CH2Cl2 at -40 °C
(HBAr4 ) [H(OEt2)2]+[B(3,5-(CF3)2C6H3)4]-),56 green
{[N3NF]Re(H)(PMe3)}BAr4 (eq 8) forms rapidly and
could be isolated as large green cubes in 76% yield.
4bH+ is diamagnetic and has 3-fold symmetry on the
does not exchange rapidly on the NMR time scale with
the terminal PH (Ht) in the phosphine ligand in 4gH+.
In the extreme one should consider a description in
which [PPh2H2]+ is bound to the d4 metal through one
P-H bond, i.e., as (η2-HPRxH3-x)+, the isoelectronic
cationic analogue of an η2-silane. Whatever the descrip-
tion, rotation of the PR1R2R3/H unit about the Re-
ligand bond axis must be facile, since the complexes
show no evidence in solution of the approximate Cs
symmetry that is found in the solid state (see below).
An example of a proton bridging between a metal and
phosphorus in a phosphine complex is known,57 but we
could find no example of what could be called an η2-
phosphonium complex in the literature. The “η2-phos-
phonium” description is viable on the basis of the fact
that 4bH+ and 4fH+ both react with ∼1 equiv of CO on
an NMR scale to give 3e and the corresponding phos-
(44) Luo, X.; Kubas, G. J .; Bryan, J . C.; Burns, C. J .; Unkefer, C. J .
J . Am. Chem. Soc. 1994, 116, 10312.
(45) Luo, X.; Kubas, G. J .; Burns, C. J .; Bryan, J . C.; Unkefer, C. J .
J . Am. Chem. Soc. 1995, 117, 1159.
(46) Schubert, U.; Muller, J .; Alt, H. G. Organometallics 1987, 6,
469.
(47) Hoyano, J . K.; Elder, M.; Graham, W. A. G. J . Am. Chem. Soc.
1969, 91, 4568.
(48) Dong, D. F.; Hoyano, J . K.; Graham, W. A. G. Can. J . Chem.
1981, 59, 1455.
(49) Karch, R.; Schubert, U. Inorg. Chim. Acta 1997, 259, 151.
(50) Graham, W. A. G.; Bennett, M. J . Chem. Eng. News 1970, 48,
75.
(51) Graham, W. A. G. J . Organomet. Chem. 1986, 300, 81.
(52) Cowie, M.; Bennett, M. J . Inorg. Chem. 1977, 16, 2321.
(53) Cowie, M.; Bennett, M. J . Inorg. Chem. 1977, 16, 2325.
(54) Luo, X.-L.; Baudry, D.; Boydell, P.; Charpin, P.; Nierlich, M.;
Ephritikhine, M.; Crabtree, R. H. Inorg. Chem. 1990, 29, 1511.
(55) Luo, X.-L.; Schulte, G. K.; Demou, P.; Crabtree, R. H. Inorg.
Chem. 1990, 29, 4268.
(56) Brookhart, M.; Grant, B.; A. F. Volpe, J . Organometallics 1992,
11, 3920.
(57) Albinati, A.; Lianza, F.; Pasquali, M.; Sommovigo, M.; Leoni,
P.; Pregosin, P. S.; Ruegger, H. Inorg. Chem. 1991, 30, 4690.

4084 Organometallics, Vol. 17, No. 18, 1998
Reid et al.
1
phonium salt, according to H, 19F, and 31P{1H} NMR.
For convenience in later discussions the proton that is
terminally bound to P will be labeled Ht, while that
which is bound to Re and interacting with P will be
labeled Hb.
Addition of 1 equiv of DBAr4 to 4g gave 4gD+, the
1H NMR spectrum of which is identical to that of 4gH+
except that the resonances for Ht and Hb each integrate
2
to ∼0.5. The H NMR spectrum (in CH2Cl2) reveals a
doublet at 6.86 ppm (J DP ) 56 Hz) and another doublet
at 3.24 ppm (J DP ) 7 Hz), showing that deuterium has
been incorporated into both the Ht (6.97 ppm in 4gH+)
and the Hb sites (3.36 ppm in 4gH+). The 31P{1H} NMR
spectrum consists of a triplet at 37.36 ppm (J DP ) 56
Hz) that can be ascribed to a Re-H-P-D complex (1H
is decoupled) and a singlet at 38.11 ppm that arises from
a Re-H-P-H complex. The singlet at 38.11 mostly
obscures another triplet with a small J DP that we ascribe
to a small amount of a Re-D-P-H complex. There is
no evidence in phosphorus NMR spectra for any Re-
D-P-D complex under the conditions used so far,
although the resonance may be weak and/or hidden.
Therefore we observe directly by 31P{1H} NMR {[N3NF]-
Re(D)(PHPh2)}+, {[N3NF]Re(H)(PDPh2)}+, and {[N3NF]-
Re(H)(PHPh2)}+. It is not known how H and D scramble
in these species. One possibility is that H and D
exchange intramolecularly between bridging and ter-
minal positions on the chemical time scale and that H+
or D+ is lost to reform 4g on a roughly equivalent time
scale. Therefore HBAr4 and [N3NF]Re(PDPh2) are
available, and at least three of the four possible isoto-
pomers can form competitively. In view of the possible
η2-phosphonium character to complexes such as 4gH+
it is worth entertaining the possibility that P-Hb
exchanges on the chemical time scale with P-Ht in an
intermediate in which two P-H bonds interact with Re
in an equivalent manner, i.e., via an “η3-phosphonium”
intermediate.
An X-ray study of 4gH+ was carried out. Two views
are shown in Figure 5, while selected bond distances
and angles are presented in Table 2. The terminal
proton (H(1)) was located and refined successfully
during the intermediate stages of refinement; the
P-H(1) bond distance (1.39(7) Å) falls within the
expected range. A second proton (H(2)) was located
between Re and P in the P-Re-N(3) plane, but became
lost repeatedly in the electron density of Re during
subsequent refinement. Therefore the Re-H(2) bond
distance was arbitrarily constrained to 1.698(5) Å and
the P-H(2) bond distance to 1.481(5) Å, both of which
are consistent with values found in other phosphine
complexes of Re hydrides.58 The PHPh2 ligand is tipped
to one side (N(4)-Re-P ) 160.2(2)°), presumably in
order to allow H(2) to lie in the same plane as the Re-
N(3) bond between P and N(3). The Re-Neq bond
distances and the Re-N(4) bond distance are slightly
longer than those in [N3NF]Re(C2H4), consistent with
more “steric pressure” in the coordination pocket where
the phosphine resides. There is no feature of the
triamido ligand in this complex that would be consistent
with one of the amido nitrogen atoms being protonated.
Interestingly, both phenyl rings of the PHPh2 ligand are
F igu r e 5. (a) Side view (ORTEP) of the structure of
{[N3NF]Re(PPh2H)(H)}+. (b) Top view of the structure of
{[N3NF]Re(PPh2H)(H)}+ (H(2) left out for clarity).
roughly parallel to two C6F5 rings ∼4 Å away and
slipped to one side. This configuration appears to be a
manifestation of interaction between C6H5 and C6F5
rings.59-61 The BAr4- anion suffered from some disorder
in the CF3 groups, several of which therefore were
refined isotropically, as explained in detail in the
Experimental Section.
Protonation of [N3NF]Re(THT) (6) with 1 equiv of
HBAr4 in cold CH2Cl2 gave 6H+ in 51% yield as green
prisms upon crystallization from mixtures of dichlo-
romethane and pentane (eq 10). The 1H NMR spectrum
of 6H+ in CD2Cl2 shows it to be C3 symmetric on the
NMR time scale with a “hydride” resonance at 3.50 ppm,
similar to the chemical shifts observed in 4cH+, 4fH+,
and 4gH+. Four multiplets between 3.26 and 1.38 ppm
are assigned to the eight THT protons, consistent with
a tetrahydrothiophene bound to Re through one lone
pair. In that circumstance exo and endo protons cannot
equilibrate on the NMR time scale, even though rotation
(58) Green, M. A.; Huffman, J . C.; Caulton, K. G. J . Am. Chem. Soc.
1982, 104, 2319.

Synthesis of Rhenium Complexes
Organometallics, Vol. 17, No. 18, 1998 4085
about the Re-S bond may be facile on the NMR time
scale, as is rotation of the entire (H)(thiophene) unit
relative to the [N3NF]3- ligand. The 13C{1H} NMR (CD2-
Cl2) spectrum shows only two THT resonances at 45.69
and 29.97 ppm, consistent with a plane of symmetry
passing through the THT ligand. The only data that
would support a proposal that the “hydride” is also
interacting to some extent with the sulfur (eq 10) is the
reaction of 6H+ with CO in CD2Cl2 to give 3e and the
(HTHT)BAr4 salt. However, this reaction also could be
viewed as a CO-induced reductive elimination reaction
from a “classical” cationic Re(V) phosphine hydride
complex.
Exp er im en ta l Section
Gen er a l P r oced u r es. All air-sensitive compounds were
manipulated under a nitrogen atmosphere in a Vacuum
Atmospheres drybox or under argon when using Schlenk
techniques. Pentane was washed with sulfuric/nitric acid (95/5
v/v), sodium bicarbonate, and then water, stored over calcium
chloride, and then distilled from sodium benzophenone ketyl
under N2. Reagent grade diethyl ether, 1,2-dimethoxyethane,
1,4-dioxane, and tetrahydrofuran were distilled from sodium;
CH2Cl2 was distilled from CaH2; acetonitrile was distilled from
P2O5. Deuterated solvents were passed through activated
alumina and vacuum-transferred to solvent storage flasks until
use. 2,2′,2′′-Tris(pentafluorophenylamino)triethylamine (H3-
[N3NF]),11 Ta(CH-t-Bu)Br3(THF)2,29 and [H(OEt2)2]+[(3,5-(CF3)2-
C6H3)4B]-56 (HBAr4) were prepared by published methods.
[NEt4]2[ReOCl5] was prepared by a method analogous to that
reported for [NMe4]2[ReOCl5].80 Phosphines and silanes were
purchased from commercial sources and used as received. THT
and pyridine were distilled from CaH2 under N2 (THT )
tetrahydrothiophene).
Protonation of [N3NF]Re(CH3CN) (7) led to formation
of the azavinylidene complex (7H+) in 75% yield as
1
green-black crystals (eq 11). The H NMR spectrum in
NMR operating frequencies and reference standards for
heteronuclei on the scale of 1H (300 MHz, SiMe4 ) 0 ppm) are
as follows: 13C (75.4 Hz, SiMe4 ) 0 ppm), 31P (121.4 MHz, 85%
H3PO4 ) 0 ppm), 19F (282.2 Hz, CFCl3 ) 0 ppm). Proton and
carbon spectra were referenced using the partially deuterated
solvent as an internal reference. Fluorine and phosphorus
spectra were referenced externally. Multiplicities in fluorine
spectra are quantified as “J ”, a pseudo coupling constant.
Chemical shifts are reported in ppm, and coupling constants
are in hertz. All spectra were acquired at ∼22 °C unless
otherwise noted.
IR spectra were recorded on a Perkin-Elmer FT-IR 16
spectrometer as Nujol mulls between KBr plates in an airtight
cell. Elemental analyses were performed on a Perkin-Elmer
PE2400 microanalyzer in our laboratory, or by Microlytics
(South Deerfield, MA) or Schwarzkopf Microanalytical Labora-
tory (Woodside, NY).
CD2Cl2 indicates that the molecule has C3 symmetry
on the NMR time scale. The methyl proton resonance
at 3.57 ppm is split into a doublet (J HH ) 7 Hz) by the
azavinylidene proton, for which a quartet is observed
at 1.54 ppm. The 13C NMR spectrum reveals a doublet
1
at 118.25 ppm with J CH ) 163 Hz that is ascribable to
the â-carbon atom. Treatment of 7H+ with DBN (DBN
) 1,5-diazabicyclo[4.3.0]non-5-ene) produced 7 rapidly.
Azavinylidene complexes in general are rare, and we
could find only one other example of an azavinylidene
complex of Re.62 We formulate the methyl azavi-
nylidene ligand in 7H+ as a linear species with the
positive charge formally located on the nitrogen atom.
In this manner the dxz or dyz orbitals are both utilized,
one to contain two electrons and one to form the RedN
covalent double bond. The 3-fold symmetry in solution
is again a consequence of equivalent dxz and dyz orbitals.
[(C6F 5NCH2CH2)2NCH2CH2NHC6F 5]Re(O)Cl (1). A CH3-
CN solution (3 mL) of H3[N3NF] (4.00 g, 6.207 mmol) and NEt3
(59) Dahl, T. Acta Chem. Scand. 1994, 48, 95.
(60) Coates, G. W.; Dunn, A. R.; Henling, L. M.; Dougherty, D. A.;
Grubbs, R. H. Angew. Chem., Int. Ed. Engl. 1997, 36, 248-251.
(61) Williams, J . H. Acc. Chem. Res. 1993, 26, 593.
(62) Chatt, J .; Dosser, R. J .; King, F.; Leigh, G. J . J . Chem. Soc.,
Dalton Trans. 1976, 2435.
(63) Stark, G. A.; Gladysz, J . A. Inorg. Chim. Acta 1998, 269, 167.
(64) Hom, R. K.; Katzenellenbogen, J . A. J . Org. Chem. 1997, 62,
6290.
(65) Archer, C. M.; Dilworth, J . R.; Griffiths, D. V.; AlJ eboori, M.
J .; Kelly, J . D.; Lu, C. Z.; Rosser, M. J .; Zheng, Y. F. J . Chem. Soc.,
Dalton Trans. 1997, 1403.
Con clu sion s
(66) Masui, C. S.; Mayer, J . M. Inorg. Chim. Acta 1996, 251, 325.
(67) Dewey, M. A.; Stark, G. A.; Gladysz, J . A. Organometallics 1996,
15, 4798.
(68) Stark, G. A.; Gladysz, J . A. Inorg. Chem. 1996, 35, 5509.
(69) Danopoulos, A. A.; Wilkinson, G.; Sweet, T. K. N.; Hursthouse,
M. B. J . Chem. Soc., Dalton Trans. 1996, 2995.
(70) Refosco, F.; Tisato, F.; Moresco, A.; Bandoli, G. J . Chem. Soc.,
Dalton Trans. 1995, 3475.
(71) Ahmet, M. T.; Coutinho, B.; Dilworth, J . R.; Miller, J . R.;
Parrott, S. J .; Zheng, Y. F.; Harman, M.; Hursthouse, M. B.; Malik, A.
J . Chem. Soc., Dalton Trans. 1995, 3041.
(72) Chi, D. Y.; R., W. S.; Katzenellenbogen, J . A. Inorg. Chem. 1995,
34, 1624.
(73) Sundermeyer, J .; Puttelik, J .; Foth, M.; Field, J . S.; Ramesar,
N. Chem. Ber. 1994, 127, 1201.
(74) Dewey, M. A.; Knight, D. A.; Arif, A.; Gladysz, J . A. Chem. Ber.-
Rec. 1992, 125, 815.
(75) Vale, M. G.; Schrock, R. R. Organometallics 1993, 12, 1140.
(76) Hoffman, D. M.; Lappas, D.; Putilina, E. Inorg. Chem. 1992,
31, 79.
(77) Vale, M. G.; Schrock, R. R. Organometallics 1991, 10, 1661.
(78) Dewey, M. A.; Arif, A. M.; Gladysz, J . A. J . Chem. Soc., Chem.
Commun. 1991, 712.
We conclude that [N3NF]Re complexes that contain
rhenium in the “mid” oxidation states (Re(III), Re(IV),
and Re(V)) are viable. This finding is somewhat sur-
prising in view of the rarity of monomeric rhenium
complexes that contain more than one amido ligand.63-79
There appears to be some preference for Re(III) com-
plexes that contain η2-bound ligands over Re(V) tau-
tomers, perhaps in part as a consequence of the ability
of the d4 metal for back-bonding, balanced with a steric
pressure to maintain the trigonal coordination pocket.
The current entry into [N3NF]Re chemistry relies on an
indirect synthesis of a low-spin d3 starting material (2)
that appears to be easily reduced. Rhenium triami-
doamine chemistry would become more amenable to
exploration if a more direct synthesis of a rhenium
starting material could be found, especially if it is
adaptable to the synthesis of triamidoamine complexes
(79) Chiu, K. W.; Wong, W.-K.; Wilkinson, G.; Galas, A. M. R.;
Hursthouse, M. B. Polyhedron 1982, 1, 37.
that contain ligands other than [N3NF]3-
.

4086 Organometallics, Vol. 17, No. 18, 1998
Reid et al.
(2.073 g, 20.480 mmol) was added to a stirred suspension of
[NEt4]2[ReOCl5] (3.973 g, 6.207 mmol) in 20 mL of CH3CN. A
rapid color change to emerald green was observed. The
solution was stirred for 30 min, and the volatile components
were removed in vacuo. The resulting green residue was
stirred into 30 mL of THF, the mixture was filtered to remove
[Et3NH]Cl, and the filtrate was concentrated in vacuo. The
product was recrystallized twice from minimal methylene
chloride by adding pentane and cooling the mixture to -40
°C; yield 643 mg (78%): 1H NMR (CD3CN) δ 2.15-4.70 (br,
NH), 3.39 and 3.53 (m, 6), 3.82 (m, 4), 4.19 (m, 2); 13C NMR
(CD3CN) δ 42.1 (br, CH2), 61.5 (br, 2 CH2), 66.4 (br, CH2), 69.7
(br, 2 CH2), 133-147 (Caryl); 19F NMR (CD3CN) δ -173.9 (tt,
J FF ) 21, 6.2), -166.8 (m), 166.4 (t, J FF ) 19), -162.4 (t, J FF
) 21), -160.3 (d, J FF ) 23), -150.4 (dd, J FF ) 23, 6.2), -149.9
(m); IR (Nujol) cm-1 3419 (NH), 910 (RedO). Anal. Calcd for
the Mg, and the reaction mixture was stirred vigorously at
room temperature overnight. All solvents were then removed
in vacuo. The olive-green residue was extracted three times
with 15 mL of THF. The extract was filtered, and excess 1,4-
dioxane was added to the combined extracts. After 24 h
[MgBr2(1,4-dioxane)]x was filtered off, THF was removed from
the filtrate in vacuo, and the residue was crystallized from
CH2Cl2/pentane at -40 °C to give small olive-green crystals;
yield 346 mg (65%): 1H NMR (THF-d8) δ 2.61 (s, 4), 3.27 (t, 6,
J ) 5.7), 3.87 (t, 6, J ) 5.7); 1H NMR (CD2Cl2) δ 2.57 (br s, 4),
3.12 (t, 6, J ) 5.7), 3.79 (t, 6, J ) 5.7); 1H NMR (CD2Cl2, -90
°C) δ 3.87 (m, 3), 3.55 (m, 3), 3.15 (m, 3), 3.04 (m, 3), 2.30 (s,
4); 19F NMR (THF-d8) δ -150.9 (m), -164.3 (t, J ) 21), -166.6
(m); 19F NMR (CD2Cl2) δ -151.1 (m), -163.2 (t, J ) 21), -165.7
(m). Anal. Calcd for C26H16N4F15Re: C, 36.50; H, 1.88; N,
6.55. Found: C, 36.70; H, 1.81; N, 6.88.
The analogous compound prepared from doubly 13C labeled
ethylene showed a non-first-order multiplet in CD2Cl2 at δ 54.7
in the 13C NMR spectrum. In the 13C{1H} NMR spectrum the
singlet showed no coupling to fluorine.
C
24H13N4ClF15ORe: C, 32.76; H, 1.49; N, 6.37; Cl, 4.03.
Found: C, 32.44; H, 1.52; N, 6.21; Cl 4.06, 4.16.
[N3NF ]ReBr (2). A solution of Ta(CH-t-Bu)Br3(THF)2 (699
mg, 1.16 mmol, 1.05 equiv) in 5 mL of toluene was slowly
added to a suspension of 1 (970 mg, 1.10 mmol) in 15 mL of
toluene at room temperature. After stirring the solution
overnight, the volume was reduced to 3 mL and the green-
brown precipitate was filtered off and washed twice with
toluene. The precipitate was recrystallized from a mixture of
methylene chloride and pentane at -40 °C to yield 814 mg of
olive-green crystals (74%): 1H NMR (CD3CN) δ -34 (v br),
-138 (v br); 19F NMR (CD3CN) δ -132.1 (br s), -151.2 (br s),
[N3NF ]Re(C3H6) (3c). One side of a two-chamber vessel
was charged with a solution of 2 (600 mg, 0.608 mmol) in 40
mL of THF. The other side was charged with Mg powder (295
mg, 12.15 mmol) and a stir bar. The solution was degassed
by three freeze-pump-thaw cycles, and propylene (excess)
was condensed into the vessel. The solution was combined
with the Mg and stirred vigorously at room temperature
overnight. All solvents were then removed in vacuo from the
olive-green mixture. The residue was extracted three times
with 15 mL of THF. The extract was filtered, and excess 1,4-
dioxane was added to the combined extracts. After 24 h
[MgBr2(1,4-dioxane)]x was filtered off, THF was removed in
vacuo, and the product was crystallized from CH2Cl2/pentane
at -40 °C. Very dark green prisms were collected and dried
in vacuo; yield 346 mg (65%): 1H NMR (CD2Cl2) δ 3.76 (m, 6,
NCH2CH2), 3.34 (m, 1, CH3CHCH2), 3.12 (d, 1, CH3CHCHH,
J HH ) 8.1), 3.05 (t, 6, NCH2CH2), 2.73 (d, 1, CH3CHCHH, J HH
) 9.3), 0.75 (d, 3, CH3CHCH2, J HH ) 5.4); 13C{1H} NMR (THF/
C6D6) δ 148-132 (m, NC6F5), 67.63, 61.56 (CH3CHCH2), 60.11
(CH3CHCH2), 56.46, 20.60 (CH3CHCH2); 19F NMR (CD2Cl2) δ
-150.20(m), -162.88 (m), -165.69 (m). Anal. Calcd for
C27H18N4F15Re: C, 37.29; H, 2.09; N, 6.44. Found: C, 37.40;
H, 1.90; N, 6.79.
[N3NF ]Re(N2) (3d ). A solution of 2 (400 mg, 0.405 mmol)
in 50 mL of THF was treated with Mg powder (295 mg, 12.15
mmol), and the mixture was stirred under N2 for 1 h, then
filtered to remove excess Mg. Excess 1,4-dioxane was added,
the solution was allowed to stand for 16 h, and [MgBr2(1,4-
dioxane)]x was filtered off. THF was removed in vacuo, and
the orange product was crystallized from CH2Cl2/pentane as
orange cubes; yield 104 mg (0.12 mmol, 72%): 1H NMR (THF-
d8) δ 3.13 (t, 6, J ) 5.6), 3.95 (t, 6, J ) 5.6); 1H NMR (CD3CN)
δ 3.07 (t, 6, J ) 5.6), 3.90 (t, 6, J ) 5.6); 19F NMR (THF-d8) δ
-152.90 (m), -164.90 (t, J ) 21), -166.59 (t, J ) 20); 19F NMR
(CD3CN) δ -152.5 (m), -164.7 (t, J ) 21), -166.5 (t, J ) 19);
13C NMR (THF-d8) δ 57.7 (t, CH2, J ) 139), 63.2 (t, CH2, J )
139), 135-145 (Caryl); 15N NMR (CD3CN) δ 281.3 and 332.0
(each a d, J NN ) 5.5); IR cm-1 2004 (νNN, vs); MS 856 (M+),
828 (-N2). Anal. Calcd for C24H12N6F15Re: C, 33.69; H, 1.41;
N, 9.82. Found: C, 33.52; H, 1.66; N, 9.48.
-156.9 (br s); MS 907 (M+). Anal. Calcd for C24H12N4BrF15
-
Re(CH2Cl2)0.5 or C49H26N8Br2F30Cl2Re2: C, 30.98; H, 1.38; N,
5.90; Br, 8.41; Cl, 3.73. Found: C, 31.20; H, 1.39; N, 6.01; Br,
8.00, 8.22; Cl, 3.81, 3.85. (The C, H, and N values are the
average of six determinations ranging from 29.73 to 31.82 in
C, 1.19 to 1.49 in H, and 5.43 to 6.19 in N.) On the basis of
the variable analyses we propose that the bulk sample is not
homogeneous, in part as a consequence of variable amounts
of dichloromethane in the crystals. (See also the discussion
of the X-ray study below.)
[N3NF ]ReH2 (3a ). Methyllithium was added to 2 (100 mg,
0.101 mmol) under dihydrogen (1 atm), and the reaction was
stirred for 12 h. Solvents were removed in vacuo, and the
residue was extracted with CH2Cl2. The reaction mixture was
filtered through Celite, and the filtrate was concentrated in
vacuo and stood at -40 °C to give 43 mg of dark red crystals
(52% yield): 1H NMR (CD3CN) δ -0.68 (s, H2), 3.10 (t, 6, J )
5.7), 3.79 (t, 6, J ) 5.7); 1H NMR (CD2Cl2) δ -0.89 (s, H2),
3.07 (t, 6, J ) 6.3), 3.80 (t, 6, J ) 5.4); 19F NMR (CD3CN) δ
-152.2 (m), -166.4 (t, J ) 21), -167.1 (m); 19F NMR (CD2Cl2)
δ -152.0 (m), -165.8 (t, J ) 22), -166.6 (m); 13C NMR (CD3-
CN) δ 57.6 (CH2), 64.3 (CH2), 136.7, 137.5, 139.8, 142.6, 144.3,
145.8 (all Caryl). Anal. Calcd for C24H14N4F15Re: C, 34.75; H,
1.70; N, 6.75. Found: C, 34.49; H, 1.86; N, 7.14.
[N3NF]Re(H)(D) (3a -d 1). The synthesis was essentially the
same as for 3a from 2 (157 mg, 0.159 mmol) in the presence
of HD gas at 1 atm for 48 h. Workup yielded 50 mg of dark
2
red crystals (38%): 1H NMR (CD3CN) δ -0.74 (t, J HD ) 17);
2
2H NMR (CH2Cl2) δ -0.98 (d, J HD ) 17). The remaining
portions of the spectra were the same as for 3a .
[N3NF ]ReD2 (3a -d 2). The synthesis was essentially the
same as for 3a from 2 (100 mg, 0.101 mmol) in the presence
of D2 gas at 1 atm for 12 h. Workup yielded 40 mg of dark
red crystals (48%): 2H NMR (THF) δ -0.43 (s). The remaining
portions of the spectra were the same as for 3a . Anal. Calcd
for C24H12D2N4F15Re: C, 34.66; H(D), 1.45; N, 6.74. Found:
C, 34.32; H(D), 1.65; N, 6.91.
[N3NF ]Re(C2H4) (3b). One side of a two-chamber vessel
was charged with a solution of 2 (400 mg, 0.608 mmol) in 40
mL of THF. The other side was charged with Mg powder (295
mg, 12.150 mmol) and a stir bar. The solution was degassed
by three freeze-pump-thaw cycles, and ethylene (excess) was
condensed into the vessel. The solution was combined with
The reaction was repeated with 168 mg of 2 in the presence
of 15N2 to yield 105 mg of the 15N-labeled compound (72%): IR
cm-1 1935 (ν
N, vs).
15 15
N
[N3NF ]Re(CO) (3e). One side of a two-chamber vessel was
charged with an olive-brown solution of 2 (500 mg, 0.506 mmol)
in 40 mL of THF. The other side was charged with Mg powder
(443 mg, 18.24 mmol) and a stir bar. The solution was
degassed by three freeze-pump-thaw cycles, and CO (excess)
was condensed into the vessel. The solution was combined
with the Mg and stirred vigorously at room temperature
overnight. All solvents were then removed in vacuo from the

Synthesis of Rhenium Complexes
Organometallics, Vol. 17, No. 18, 1998 4087
pale yellow mixture. The residue was extracted with 3 × 15
mL of THF. The extract was filtered, and excess 1,4-dioxane
was added to the combined extracts. After 24 h [MgBr2(1,4-
dioxane)]x was filtered off, THF was removed in vacuo, and
the product was crystallized from CH2Cl2/pentane at -40 °C
as pale yellow blocks; yield 352 mg (81%). The reaction was
repeated with 50 mg of 2 under 13CO to give the 13C-labeled
product in the same yield: 1H NMR (CD2Cl2) δ 3.16 (t, 6, J )
5.7), 3.95 (t, 6, J ) 5.4); 19F NMR (CD2Cl2) δ -152.0 (m),
-163.7 (t, J ) 22), -165.6 (m); 13C NMR (THF-d8) δ 55.5 (CH2),
62.6 (CH2), 135-145 (Caryl); 13C NMR (CD2Cl2) δ 199.5 (13CO);
IR (DME) cm-1 1875 cm-1 (CdO), 1830 (13CdO). Anal. Calcd
for C25H12N4F15ORe: C, 35.10; H, 1.41; N, 6.55. Found: C,
35.25; H, 1.29; N, 6.50.
[N3NF]Re(P H3) (4a ). One side of a two-chamber vessel was
charged with a solution of 2 (485 mg, 0.491 mmol) in 50 mL of
THF. The other side was charged with Mg powder (358 mg,
14.74 mmol) and a stir bar. The solution was degassed by
three freeze-pump-thaw cycles, and PH3 (∼10 equiv) was
condensed into the vessel. The mixture was warmed to room
temperature, combined with the Mg, and stirred vigorously
at room temperature for 6 h. Solvents were then removed from
the red mixture in vacuo. Excess PH3 was decomposed by
venting the trap to a saturated aqueous solution of CuSO4.
The red residue was extracted three times with 15 mL of THF.
The extract was filtered and excess 1,4-dioxane was added to
the combined extracts. After 16 h [MgBr2(1,4-dioxane)]x was
filtered off, THF was removed in vacuo, and the product was
crystallized from CH2Cl2/pentane at -40 °C. Light red crystals
were collected in two crops; yield 171 mg (40%): 1H NMR (CD2-
Cl2) δ 5.38 (d, 3, J PH ) 305) 3.79 (t, 6, NCH2CH2), 3.00 (t, 6,
NCH2CH2); 13C{1H} NMR (CD2Cl2) δ 148-132 (m, NC6F5),
65.49, 57.09; 19F NMR (CD2Cl2) δ -152.50 (m), -165.34 (t, J FF
) 21), -166.16 (m); 31P{1H} NMR (CD2Cl2) δ -138.18 (s). Anal.
Calcd for C24H15N4F15PRe: C, 33.46; H, 1.75; N, 6.50. Found:
C, 33.11; H, 1.55; N, 6.38.
PMe2Ph (168 mg, 1.215 mmol) in 50 mL of THF, and Mg
powder (295 mg, 12.15 mmol); yield 165 mg (42%) of a red-
orange microcrystalline solid: 1H NMR (CD2Cl2) δ 7.08-6.77
(m, 5, Ph), 3.67 (t, 6, NCH2CH2), 2.87 (t, 6, NCH2CH2), 0.54
(d, 9, J PH ) 6.4, PCH3); 13C{1H} NMR (CD2Cl2) δ 148-135 (m,
NC6F5), 129.71 (d, Ph, J PC ) 10.9), 128.49 (s, Ph), 127.53 (d,
Ph, J PC ) 8.4), 67.40, 56.14, 23.28 (d, CH3, J PC ) 27.8); 19F
NMR (CD2Cl2) δ -149.25 (m), -164.70 (t, J FF ) 21), -165.83
(m); 31P{1H} NMR (CD2Cl2) δ -30.91 (s). Anal. Calcd for
C
32H23N4F15PRe: C, 39.80; H, 2.40; N, 5.80. Found: C, 39.36;
H, 2.16; N, 5.64.
[N3NF ]Re(P MeP h 2) (4f). The procedure was analogous to
that used to prepare 4a employing 2 (400 mg, 0.405 mmol),
PMePh2 (263 mg, 1.215 mmol) in 50 mL of THF, and Mg
powder (295 mg, 12.15 mmol); yield 349 mg (84%) of red
crystals: 1H NMR (CD2Cl2) δ 6.48-7.10 (m, 10, Ph), 3.66 (t,
6, NCH2CH2), 2.87 (t, 6, NCH2CH2), 1.02 (d, 3, J PH ) 6.0,
PCH3); 13C{1H} NMR (CD2Cl2) δ 148-132 (m, NC6F5), 130.87
(d, J PC ) 9.5), 128.46, 127.81 (d, J PC ) 7.8), 68.32, 56.18, 22.10
(d, J PC ) 28.3, PCH3); 19F NMR (CD2Cl2) δ -148.76 (bs),
-164.62 (t, J FF ) 21), -165.48 (m); 31P{1H} NMR (CD2Cl2) δ
-15.19 (s). Anal. Calcd for C37H25N4F15PRe: C, 43.24; H, 2.45;
N, 5.45. Found: C, 43.19, H, 2.45; N, 5.45.
[N3NF ]Re(P HP h 2) (4g). The procedure was analogous to
that used to prepare 4a employing 2 (400 mg, 0.405 mmol),
PHPh2 (226 mg, 1.215 mmol) in 50 mL of THF, and Mg powder
(295 mg, 12.15 mmol); yield 376 mg (91%) of orange cubes:
1H NMR (CD2Cl2) δ 9.40 (d, 1, J PH ) 319, PH), 6.45-7.12 (m,
10, Ph), 3.73 (t, 6, NCH2CH2), 2.96 (t, 6, NCH2CH2); 13C{1H}
NMR (CD2Cl2) δ 142-135 (m, NC6F5), 136.67, 131.92, 128.81,
128.33, 67.39, 56.63; 19F NMR (CD2Cl2) δ -150.38 (m), -165.00
(t, J FF ) 21), -165.53 (m); 31P{1H} NMR (CD2Cl2) δ 1.24 (s).
Anal. Calcd for C36H23N4F15PRe: C, 42.65; H, 2.29; N, 5.53.
Found: C, 42.71; H, 2.48; N, 5.36.
[N3NF ]Re(p yr id in e) (5). The procedure was analogous to
that used to prepare 4a employing 2 (400 mg, 0.405 mmol)
and pyridine (160 mg, 2.03 mmol) in 50 mL of THF, and a
solution of MeLi (1.4 M in Et2O, 0.61 mL, 0.851 mmol) in 20
mL of diethyl ether. The residue was extracted three times
with 15 mL of CH2Cl2 instead of THF; yield 179 mg (48%):
1H NMR (CD2Cl2) δ 6.54 (d, 2, Ph), 6.15 (m, 2, Ph), 5.92 (m, 1,
Ph), 3.64 (t, 6, NCH2CH2), 2.73 (t, 6, NCH2CH2); 13C{1H} NMR
(CD2Cl2) δ 158.03, 142-135 (m, NC6F5), 130.59, 122.39, 70.09,
61.27; 19F NMR (CD2Cl2) δ -150.57 (m), -165.47 (t, J FF ) 21),
-166.59 (m). Anal. Calcd for C29H17N5F15Re: C, 38.42; H,
1.89; N, 7.72. Found: C, 38.28; H, 2.02; N, 7.84.
[N3NF ]Re(THT) (6). The procedure was analogous to that
used to prepare 4a employing 2 (400 mg, 0.405 mmol) and THT
(179 mg, 2.03 mmol) in 50 mL of THF and Mg powder (295
mg, 12.15 mmol); yield 252 mg (68%): 1H NMR (CD2Cl2) δ 3.62
(t, 6, NCH2CH2), 2.80 (t, 6, NCH2CH2), 1.23 (br s, 8, THT);
13C{1H} NMR (CD2Cl2) δ 148-132 (m, NC6F5), 68.14, 59.82,
45.69, 29.97; 19F NMR (CD2Cl2) δ -150.68 (m), -165.10 (t, J FF
) 21), -165.91 (m). Anal. Calcd for C28H20N4F15ReS: C,
36.73; H, 2.20; N, 6.12. Found: C, 36.59; H, 2.29; N, 6.17.
[N3NF ]Re(NCMe) (7). The procedure was analogous to
that used to prepare 4a employing 2 (2.00 g, 2.03 mmol) and
MeCN (249 mg, 6.08 mmol) in 40 mL of THF and Mg powder
(1.48 g, 60.78 mmol); yield 1.08 g (61%) of olive-green prisms:
1H NMR (CD2Cl2) δ 3.80 (t, 6, NCH2CH2), 3.51 (s, 3, NCCH3),
2.84 (t, 6, NCH2CH2); 13C{1H} NMR (CD2Cl2) δ 148-132 (m,
NC6F5), 130.18, 65.97, 60.20, 0.93; 19F NMR (CD2Cl2) δ -152.62
(m), -167.75 (m); IR (Nujol) 2151.8 cm-1 (CtN). Anal. Calcd
for C26H15N5F15Re: C, 35.95; H, 1.74; N, 8.06. Found: C,
35.81; H, 1.67; N, 8.26.
[N3NF ]Re(P Me3) (4b). The procedure was analogous to
that used to prepare 4a employing 2 (400 mg, 0.405 mmol),
PMe3 (123 mg, 1.620 mmol) in 50 mL of THF, and Mg powder
(295 mg, 12.15 mmol); yield 274 mg, 75%: 1H NMR (CD2Cl2)
δ 3.63 (t, 6, NCH2CH2), 2.82 (t, 6, NCH2CH2), 0.14 (d, 9, J PH
)
5.4, PCH3); 13C{1H} NMR (CD2Cl2) δ 148-132 (m, NC6F5),
67.21, 56.60, 23.27 (d, J PC ) 27, PCH3); 19F NMR (CD2Cl2) δ
-150.74 (m), -164.85 (t, J FF ) 21), -165.90 (m); 31P{1H} NMR
(CD2Cl2) δ -45.16 (s). Anal. Calcd for C27H21N4F15PRe: C,
35.89; H, 2.34; N, 6.20. Found: C, 35.83; H, 2.37; N, 6.20.
[N3NF ]Re(P Et3) (4c). The procedure was analogous to that
used to prepare 4a employing 2 (500 mg, 0.506 mmol), PEt3
(90 mg, 0.760 mmol) in 40 mL of THF, and Mg powder (369
mg, 15.18 mmol); yield 313 mg (65%) of dark red crystals: 1H
NMR (CD2Cl2) δ 3.60 (t, 6, NCH2CH2), 2.77 (t, 6, NCH2CH2),
0.43 (m, 15, PEt3); 13C{1H} NMR (THF/C6D6) δ 148-132 (m,
NC6F5), 68.58, 56.29, 24.70 (d, CH2, J CP ) 24.8), 8.51; 19F NMR
(CD2Cl2) δ -149.00 (b), -164.53 (t, J FF ) 21), -166.18 (m);
31P{1H}NMR (CD2Cl2)δ-24.92 (s). Anal. Calcd for C30H27N4F15
-
PRe: C, 38.10; H, 2.88; N, 5.92. Found: C, 38.35; H, 2.56; N,
5.94.
[N3NF ]Re[P (i-P r )3] (4d ). The procedure was analogous to
that used to prepare 4a employing 2 (600 mg, 0.608 mmol),
P(i-Pr)3 (146 mg, 0.912 mmol) in 50 mL of THF, and Mg
powder (443 mg, 18.23 mmol); yield 408 mg (68%) of dark
green crystals: 1H NMR (CD2Cl2) δ 3.53 (t, 6, NCH2CH2), 2.67
(t, 6, NCH2CH2), 0.81 (dd, 18, J HH ) 7, J PH ) 12, PCHCH3),
0.48 (m, 3, PCHCH3); 13C{1H} NMR (THF/C6D6) δ 145-135
(m, NC6F5), 71.41, 55.55, 33.19 (d, CHP, J PC ) 18.4), 19.69;
19F NMR (CD2Cl2) δ -146.52 (m), -163.69 (t, J FF ) 21),
-165.97 (m); 31P{1H} NMR (CD2Cl2) δ -11.56 (s). Anal. Calcd
for C33H33N4F15PRe: C, 40.13; H, 3.37; N, 5.67. Found: C,
40.21; H, 3.33; N, 5.55.
[N3NF ]Re(H2SiEt2) (8a ). The procedure was analogous to
that used to prepare 4a employing 2 (2.00 g, 2.03 mmol) and
Et2SiH2 (536 mg, 6.08 mmol) in 50 mL of THF and Mg powder
(1.48 g, 60.8 mmol). The product was crystallized from DME/
pentane at -40 °C; yield 911 mg (49%) of orange-red microc-
rystals: 1H NMR (THF-d8) δ 5.83 (s, 1, SiH, J SiH ) 186), 3.80
[N3NF ]Re(P Me2P h ) (4e). The procedure was analogous to
that used to prepare 4a employing 2 (400 mg, 0.405 mmol),

4088 Organometallics, Vol. 17, No. 18, 1998
Reid et al.
(t, 6, NCH2CH2), 3.11 (t, 6, NCH2CH2), 0.32 (t, 9, SiMe3), -0.01
(m, 2, SiCH2), -5.85 (s, 1, Si-H, J SiH ) 44); 13C{1H} NMR
(THF/C6D6) δ 148-132 (m, NC6F5), 66.53, 55.93, 10.16 (s,
SiCH2CH3), 7.38 (s, SiCH2CH3); 19F NMR (CD2Cl2) δ -149.58
(b), -164.97 (t, J FF ) 21), -166.85 (m). Anal. Calcd for
mL of CH2Cl2 at -40 °C; yield 243 mg (66%) of red-brown
blocks: 1H NMR (CD2Cl2) δ 7.73 (bs, 8, o-H), 7.56 (s, 4, p-H),
6.73-7.40 (m, 10, Ph), 4.72 (d, 1, J PH ) 62, Re-H), 4.04 (t, 6,
NCH2CH2), 3.48 (t, 6, NCH2CH2), 2.01 (d, 3, J PH ) 9.6, PCH3);
13C{1H} NMR (CD2Cl2) δ 148-132 (m, NC6F5), 162.50 (q, Cipso
,
C
28H24N4F15ReSi: C, 36.72; H, 2.64; N, 6.12. Found: C, 36.53;
J BC ) 50), 145-136 (m, NC6F5), 136.01 (s, Co), 133.40 (d, Cipso
,
H, 2.47; N, 6.11.
J PC ) 2), 131.27 (s, Co), 130.40 (m, Cm), 129.62 (q, Cm, J CF
)
29), 125.22 (q, CF3, J CF ) 272), 118.22 (s, Cp), 118.08 (s, Cp),
67.45, 53.51, 23.69 (d, PCH3, J PC ) 48); 19F NMR (CD2Cl2) δ
-63.24 (m), -146.03 (bs), -155.71 (t, J FF ) 21), -161.26 (m);
[N3NF ]Re(H3SiP h ) (8b). The procedure was analogous to
that used to prepare 4a employing 2 (500 mg, 0.506 mmol)
and PhSiH3 (110 mg, 1.013 mmol) in 30 mL of THF and Mg
powder (369 mg, 15.18 mmol). Orange-red needles were
obtained from DME/pentane at -40 °C; yield 431 mg (89%):
1H NMR (CD3CN) δ 7.07 (m, 1, Ph), 6.94 (m, 2, Ph), 6.60 (m,
2, Ph), 5.00 (s, 2, SiH, J SiH ) 183), 3.81 (t, 6, NCH2CH2), 3.15
(t, 6, NCH2CH2), -3.86 (s, 1, Si-H-Re, J SiH ) 38); 13C{1H}
NMR (THF/C6D6) 145-138 (m, NC6F5), 134.61, 128.47, 127.53,
65.94, 55.94; 19F NMR (CD3CN) δ -150.37 (m), -163.92 (t,
J FF ) 21), -166.36 (m). Anal. Calcd for C30H20N4F15ReSi: C,
38.51; H, 2.15 N, 5.99. Found: C, 38.57 H, 2.14; N, 6.06.
[D(OEt2)2]+BAr 4-. A 250 mL round-bottom Schlenk flask
was charged with a stir bar and a solution of NaBAr4 (5.00 g,
5.64 mmol) in 125 mL of diethyl ether. The flask was
transferred to a high-vacuum line and cooled to -196 °C, and
a slight excess of DCl was condensed into the flask. The
mixture was stirred and warmed to room temperature. The
mixture was allowed to stir for 5 min, and then the flask was
pumped into the drybox. The reaction mixture was chilled to
-40 °C, and the reaction solution was filtered through a bed
of Celite. The pale yellow filtrate was concentrated, and the
product was isolated as white crystals upon addition of pentane
and cooling to -40 °C overnight; yield 4.10 g (72%): 1H NMR
(CD2Cl2) δ 7.73 (b, 8, o-H), 7.58 (br, 4, p-H), 3.85 (q, 8, OCH2-
31P{1H} NMR (CD2Cl2) δ 23.74. Anal. Calcd for C69H38N4BF39
-
PRe: C, 43.80; H, 2.02; N, 2.96. Found: C, 43.88; H, 1.70; N,
2.82.
{[N3NF ]Re(H)(P HP h 2)}+BAr 4- (4gH+). This complex was
prepared in a manner analogous to that employed to prepare
4bH+ from 4g (200 mg, 0.195 mmol) in 15 mL of CH2Cl2 at
-40 °C and [H(OEt2)2]BAr4 (197 mg, 0.195 mmol) in 2 mL of
CH2Cl2 at -40 °C; yield 257 mg of coral-colored needles
(69%): 1H NMR (CD2Cl2) δ 7.73 (bs, 8, o-H), 7.56 (s, 4, p-H),
6.85-7.40 (m, 10, Ph), 6.97 (dd, 1, J HH ) 15, J PH ) 375, Re-
H), 4.05 (t, 6, NCH2CH2), 3.57 (t, 6, NCH2CH2), 3.36 (dd, 1,
J HH ) 15, J PH ) 54, Re-H); 13C{1H} NMR (CD2Cl2) δ 162.42
(q, Cipso, J BC ) 50), 145-136 (m, NC6F5), 135.44 (s, Co), 133.00
(d, Cipso, J PC ) 2), 131.22 (s, Co), 130.40 (m, Cm), 129.49 (q, Cm,
J CF ) 29), 125.24 (q, CF3, J CF ) 272), 118.04 (s, Cp), 118.00 (s,
Cp), 66.90, 55.51; 19F NMR (CD2Cl2) δ -63.28 (m), -149.04
(s), -155.41 (t, J FF ) 21), -161.18 (m); 31P{1H} NMR (CD2-
Cl2) δ 38.11. Anal. Calcd for C68H36N4BF39PRe: C, 43.49; H,
1.93; N, 2.98. Found: C, 43.70; H, 1.84; N, 2.60.
{[N3NF ]Re(D)(P HP h 2)}+BAr 4- (4gD+). This complex was
prepared in a manner identical to that used to prepare 4gH+
from 4g (503 mg, 0.496 mmol) and [D(OEt2)2]BAr4 (503 mg,
0.496 mmol); yield 691 mg (74%): 2H NMR (CH2Cl2) δ 6.86
(d, Re-D-P, J DP ) 56), 3.24 (d, P-D, J DP ) 7); 31P{1H} NMR
δ 38.11 (s), 38.10 (t, obscured by singlet at 38.11), 37.36 (t,
J DP ) 56).
2
CH3), 1.31 (t, 12, OCH2CH3); H NMR (CH2Cl2) δ 12.84 (br).
{[N3NF ]R e(H )(P Me3)}+BAr 4 (4b H+). A solution of 4b
-
(200 mg, 0.221 mmol) in 15 mL of CH2Cl2 at -40 °C was
stirred, and a solution of [(3,5-(CF3)2C6H3)4B]-[H(OEt2)2]+ (224
mg, 0.221 mmol) in 2 mL of CH2Cl2 at -40 °C was added
dropwise over a period of 1 min. The reaction mixture turned
dark red-green immediately and gradually faded to green after
being stirred overnight. The solution was concentrated, and
the product was crystallized from CH2Cl2/pentane at -40 °C
as emerald green blocks; yield 281 mg (76%): 1H NMR (CD2-
Cl2) δ 7.78 (s, 8, o-H), 7.59 (s, 4, p-H), 4.00 (t, 6, NCH2CH2),
3.43 (t, 6, NCH2CH2), 1.01 (d, 9, J PH ) 9, PMe), 0.26 (d, 1, J PH
-
{[N3NF ]Re(H)(THT)}+BAr 4 (6H+). This complex was
prepared in a manner analogous to that employed to prepare
4bH+ from 6 (200 mg, 0.218 mmol) in 15 mL of CH2Cl2 at -40
°C and [H(OEt2)2]BAr4 (221 mg, 0.218 mmol) in 2 mL of CH2-
Cl2 at -40 °C; yield 193 mg of green prisms (51%): 1H NMR
(CD2Cl2) δ 7.75 (s, 8, Ho), 7.57 (s, 4, Hp), 3.81 (t, 6, NCH2CH2),
3.50 (s, 1, Re-H), 3.26 (t, 6, NCH2CH2), 2.92 (m, 2, THT), 2.48
(m, 2, THT), 2.08 (m, 2, THT), 1.38 (m, 2, THT); 13C{1H} NMR
(CD2Cl2) δ 162.49 (q, Cipso, J BC ) 50), 146-136 (m, NC6F5),
) 7, Re-H); 13C{1H} NMR (CD2Cl2) δ 162.44 (q, Cipso, J BC
50), 146-136 (m, NC6F5), 135.48 (s, Co), 129.27 (q, Cm, J CF
)
)
135.52 (s, Co), 129.57 (q, Cm, J CF ) 32), 125.32 (q, CF3, J CF
)
32), 125.25 (q, CF3, J CF ) 272), 118.11 (s, Cp), 66.98, 55.23,
21.57 (d, PCH3, J PC ) 41); 19F NMR (CD2Cl2) δ -63.32 (m),
-148.83 (m), -154.94 (t, J FF ) 21), -160.55 (m); 31P{1H} NMR
(CD2Cl2) δ -14.88. Anal. Calcd for C59H34N4BF39PRe: C,
40.08; H, 1.94; N, 3.17. Found: C, 40.21; H, 1.85; N, 3.11.
272), 118.15 (s, Cp), 65.19 (THT), 61.80, 56.15, 29.28 (THT);
19F NMR (CD2Cl2) δ -63.27 (m), -148.78 (m), -155.30 (t, J FF
) 21), -159.92 (m); 31P{1H} NMR (CD2Cl2) δ -14.88. Anal.
Calcd for
C60H33N4BF39ReS: C, 40.49; H, 1.87; N, 3.15.
Found: C, 40.49; H, 1.57; N, 3.21.
{[N3NF ]Re(H)(P Et3)}+BAr 4 (4cH+). This complex was
{[N3NF ]Re[NC(H)Me]}+BAr 4 (7H+). This complex was
-
-
prepared in a manner analogous to that employed to prepare
4bH+ from 4c (195 mg, 0.206 mmol) in 10 mL of CH2Cl2 at
-40 °C and [H(OEt2)2]BAr4 (209 mg, 0.206 mmol) in 2 mL of
CH2Cl2 at -40 °C; yield 323 mg of amber crystals (97%): 1H
NMR (CD2Cl2) δ 7.72 (s, 8, o-H), 7.57 (s, 4, p-H), 3.98 (t, 6,
NCH2CH2), 3.71 (d, 1, Re-H-P, J PH ) 64), 3.40 (t, 6,
NCH2CH2), 1.19 (m, 6, J PH ) 9, PCH2CH3), 0.66 (m, 9,
prepared in a manner analogous to that employed to prepare
4bH+ from 7 (200 mg, 0.230 mmol) in 15 mL of CH2Cl2 at -40
°C and [H(OEt2)2]BAr4 (233 mg, 0.230 mmol) in 2 mL of CH2-
Cl2 at -40 °C. The solution was allowed to stir for 2 h; yield
299 mg of green-black crystals (75%): 1H NMR (CD2Cl2) δ 7.77
(s, 8, o-H), 7.58 (s, 4, p-H), 4.18 (t, 6, NCH2CH2), 3.77 (t, 6,
NCH2CH2), 3.57 (d, 3, J HH ) 6.6, NCCH3(H)), 1.54 (q, 1, J HH
) 6, NCCH3(H)); 13C{1H} NMR (CD2Cl2) δ 162.62 (q, Cipso, J BC
) 50), 146-136 (m, NC6F5), 135.66 (s, Co), 129.73 (q, Cm, J CF
PCH2CH3); 13C{1H} NMR (THF/C6D6) δ 162.83 (q, Cipso, J BC
50), 146-136 (m, NC6F5), 135.60 (s, Co), 130.00 (q, Cm, J CF
)
)
1
32), 125.50 (q, CF3, J CF ) 272), 118.15 (s, Cp), 67.45, 54.76,
) 32), 125.25 (q, CF3, J CF ) 272), 118.25 (s, NCCH3(H), J CH
21.61 (d, PCH2CH3, J PC ) 35), 8.35 (d, PCH2CH3, J PC ) 6); 19
F
) 163), 66.34, 57.78, -0.61 (s, NCCH3(H)); 19F NMR (CD2Cl2)
δ -63.22 (m), -149.14 (m), -154.67 (t, J FF ) 21), -160.98 (m).
Anal. Calcd for C58H27N5BF39Re: C, 40.23; H, 1.57; N, 4.04.
Found: C, 40.51; H, 1.54; N, 4.11.
Liga n d Su bstitu tion Exp er im en ts. In a typical experi-
ment, 20 mg of a given complex was dissolved in CH2Cl2 (0.7
mL) at 22 °C with ∼3 equiv of donor ligand in an NMR tube,
and the mixture was allowed to stand at ca. 22 °C for 1-3
days. Reactions were followed by 19F NMR. For reactions at
NMR (CD2Cl2) δ -63.30 (m), -147.55 (m), -154.84 (t, J FF
)
21), -160.77 (m); 31P{1H} NMR (CD2Cl2) δ 23.02. Anal. Calcd
for C62H40N4BF39PRe: C, 41.14; H, 2.23; N, 3.10. Found: C,
40.81; H, 2.12; N, 3.05.
{[N3NF ]Re(H)(P MeP h 2)}+BAr 4 (4fH+). This complex
-
was prepared in a manner analogous to that employed to
prepare 4bH+ from 4f (200 mg, 0.195 mmol) in 15 mL of CH2-
Cl2 at -40 °C and [H(OEt2)2]BAr4 (197 mg, 0.195 mmol) in 2

Synthesis of Rhenium Complexes
Organometallics, Vol. 17, No. 18, 1998 4089
high temperature the NMR tube was flame-sealed and kept
at 85-120 °C for hours to days. Reactions were monitored by
19F NMR.
most consistent with the presence of ∼1 Br and ∼1 Cl per Re
in the bulk sample. Refinement of that atom as a 100%
bromide resulted in an equivalent isotropic thermal parameter,
U, of 0.20 Å2. Refinement as a 100% chloride gave a U of 0.009
Å2. Hence, the site cannot be occupied by either bromide or
chloride alone. A mixture of the two (60% bromide, 40%
chloride) proved satisfactory. In the final refinement identical
coordinate and anisotropic thermal parameters were imposed
on each of the atoms. If this crystal were representative of
the bulk sample and contained 0.5 equiv of dichloromethane,
the elemental analyses should be different from the average
observed in six analyses of bulk material. Therefore we
surmise that the bulk sample is not homogeneous and that
the particular crystal chosen for this X-ray study was not
representative of the average elemental composition.
Rea ction of 4bH+ w ith CO. A 120 mL pressure vessel
was charged with a green solution of 4bH+ (500 mg, 0.299
mmol) in 25 mL of CH2Cl2. The solution was frozen to 77 K,
the headspace evacuated, and ∼1 equiv CO was introduced to
the vessel, which was then sealed. Upon warming the vessel
to room temperature, the green slurry rapidly turned to a
golden-brown solution. After 2 days the golden-brown solution
was concentrated in vacuo. Upon addition of pentane to the
mixture and cooling it to -40 °C, a green solid precipitated,
which contained (HPMe3)BAr4 as the only identifiable spe-
cies: 1H NMR (CD2Cl2) δ 7.91, 7.66, 6.22 (m, HP(CH3)3), 1.95
(dd, HP(CH3)3); 19F NMR (CD2Cl2) δ -62.86 (s).
Rea ction s of 4bH+, 4fH+, a n d 6H+ w ith CO on a n NMR
Sca le. In a typical experiment, an NMR tube containing ∼20
mg of the complex dissolved in 0.7 mL of CD2Cl2 was charged
with ∼1 equiv CO and then flame-sealed. Reactions were
The crystal of 2 in this X-ray study also suffers from disorder
in the ligand backbone; the carbon atoms attached to N(4) are
found in the two positions that give a “clockwise” or “counter-
clockwise” twist to the NC3 unit. This disorder was effectively
modeled, and the disordered atoms appear in pairs as C(2)
and C(2)′, C(4) and C(4)′, and C(6) and C(6)′ in the Supporting
Information. (Only C(2), C(4), and C(6) are shown in Figure
2.) Owing to the disorder these carbon atoms were refined
isotropically and the hydrogen atoms attached to them were
not included in the final refinement, while all other non-
hydrogen atoms were refined as anisotropic scatterers.
1
followed by H, 19F, and 31P{1H} NMR. After 1 day, reaction
mixtures contained only 3e and the (HL)BAr4 salt.
Solid -Sta te Ma gn etic Su scep tibility Mea su r em en ts.
SQUID experiments were performed at 5 KG on a Quantum
Design 5.5 T instrument running MSRP2 software. Samples
were prepared in an N2-filled drybox. A gelatin capsule and
a 2.2 by 1.9 cm piece of Parafilm were weighed. The capsule
was then loaded with the sample, and the Parafilm was folded
and packed on top using plastic tongs. The capsule was closed
and weighed again to determine the sample mass. It was then
suspended in a straw. The straw was placed in a plastic bottle
with a screw cap, and the bottle was tightly sealed. At the
instrument the straw was quickly attached to the sample rod
and transferred to the helium atmosphere. Measurements
were taken in 1 deg intervals from 5 to 10 K, 2 deg intervals
from 12 to 20 K, 3 deg intervals from 23 to 50 K, 5 deg intervals
from 55 to 100 K, 10 deg intervals from 110 to 200 K, and 20
deg intervals from 220 to 300 K. A background measurement
of an empty gel capsule, Parafilm square, and straw was taken
over the entire temperature range and subtracted from the
experimental values at each temperature. The susceptibility
at each temperature also was corrected for the diamagnetic
contribution by the ligands using Pascal’s constants.
The solution of 3b was carried out in a manner analogous
to that described for 2. All atoms were refined as anisotropic
scatterers, and hydrogen atoms were placed in calculated
positions. Large residual electron density resides within 1.0
Å of the Re atom.
The solution of 4gH+ was carried out in a manner analogous
to that described for 2. Two problems arose during refine-
ment: a high degree of disorder in certain CF3 groups in the
anion, and the treatement of hydrogen atoms in the PPh2H2
ligand. The former problem was solved fortuitously by the fact
that one CF3 group was well-behaved. All disordered CF3
groups were refined isotropically on the basis of the CF3 group
that was well-behaved (C(65), F(651), F(652), and F(653)). The
latter problem was more significant. In the difference Fourier
maps two peaks of approximately 1.1 e/Å3 were located within
1.5 Å of the phosphorus atom. Free refinement of these as
hydrogen atoms resulted in the encroachment of one, H(2),
upon the rhenium atom. The final model is one based upon
both hydrogen atoms fixed at ∼1.4 Å from the phosphorus
atom.
X-r a y Str u ctu r a l Deter m in a tion of 2, 3b, a n d 4gH +. A
Siemens SMART/CCD area detection system based upon a
three-circle platform goniometer was used for all data collec-
tion. Unit cell parameters were obtained from 45 0.3° omega
scans. Final cell parameters were determined by least-squares
refinement of the largest (>10σ(I)) reflections measured with
the SAINT package. Crystal data are given in Table 1. No
absorption corrections were applied.
The initial solution of 2 was carried out by direct methods,
and the positions of the remaining non-hydrogen atoms were
ascertained from difference Fourier calculations. During the
course of solution and refinement a number of issues arose.
First, a considerable amount of electron density was located
away from the ReN4 core. The electron density (largest
electron density of 4.6 e/Å3) was most easily attributed to
lattice methylene chloride. However, the position was only
partially occupied by solvent, and the solvent was found to
take up two positions in the asymmetric unit. In addition the
solvent was disordered. The structure based upon the best
modeling of the solvent gave a residual (R1) of 0.085. Upon
removing the solvent contribution using SQUEEZE,81 the
residual was reduced to the reported value (R1 ) 0.041).
Distances and angles and their associated esd’s were es-
sentially unchanged within the main halide derivative.
The second problem that arose was the identity of the moiety
in the trigonal coordination pocket. Elemental analyses were
Ack n ow led gm en t. R.R.S. thanks the National
Institutes of Health (GM 31978) for research support,
S.M.R. thanks Michael Fickes for his assistance in
magnetization transfer experiments, and B.N. thanks
the Deutsches Forschungs Gemeinschaft for partial
support through a postdoctoral fellowship.
Su p p or tin g In for m a tion Ava ila ble: Labeled ORTEP
drawing, crystal data and structure refinement, atomic coor-
dinates, bond lengths and angles, and anisotropic displacement
parameters for [N3NF]ReX, [N3NF]Re(C2H4), and {[N3NF]Re-
(H)(PHPh2)}BAr4 (34 pages). This material is contained in
many libraries on microfiche, immediately follows this article
in the microfilm edition of this journal, and can be ordered
from the ACS; see any current masthead page for ordering
information.
(80) Grove, D. E.; Wilkinson, G. J . Chem. Soc. (A) 1966, 1224.
(81) van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990, A46, 194.
OM980220+