Selective ion pairing in [Ir(bipy)H2(PRPh2)2]A (A = PF6, BF4, CF3SO3, BPh4, R = Me, Ph): Experimental identification and theoretical understanding

Article abstract of DOI:10.1021/om010015l

NMR studies of the title compounds show that selective tight ion pairing occurs with the anion binding site being located in an intuitively unexpected region of the cation: on the side of the bipyridyl ligand remote from the metal instead of being closest to the metal near the MH₂ group. The interaction specificity falls off with increasing anion size. ONIOM (QM/MM) calculations represent the cation structure very well, and natural population analysis calculations identify the predominant location of the positive charge at the bipyridyl ring carbons that take part in the inter-ring C-C bond (C-2 and C-2′). This is fully consistent with tight ion pairing near these and the adjacent carbons, C-3 and C-3′. The electrostatic potential calculated for the cation confirms the location of the preferred binding site.

Full text of DOI:10.1021/om010015l

Organometallics 2001, 20, 2367-2373
2367
Selective Ion P a ir in g in [Ir (bip y)H2(P RP h 2)2]A (A ) P F 6,
BF 4, CF 3SO3, BP h 4, R ) Me, P h ): Exp er im en ta l
Id en tifica tion a n d Th eor etica l Un d er sta n d in g
Alceo Macchioni,*,† Cristiano Zuccaccia,† Eric Clot,*,‡ Karin Gruet,§ and
Robert H. Crabtree*,§
Dipartimento di Chimica, Universita` di Perugia, Via Elce di Sotto 8, 06123 Perugia, Italy,
L.S.D.S.M.S. (UMR 5636), Case Courrier 14, Universite´ de Montpellier II, 34095 Montpellier
Cedex 5, France, and Department of Chemistry, Yale University, 225 Prospect Street,
New Haven, Connecticut 06520-8107
Received J anuary 4, 2001
NMR studies of the title compounds show that selective tight ion pairing occurs with the
anion binding site being located in an intuitively unexpected region of the cation: on the
side of the bipyridyl ligand remote from the metal instead of being closest to the metal near
the MH2 group. The interaction specificity falls off with increasing anion size. ONIOM (QM/
MM) calculations represent the cation structure very well, and natural population analysis
calculations identify the predominant location of the positive charge at the bipyridyl ring
carbons that take part in the inter-ring C-C bond (C-2 and C-2′). This is fully consistent
with tight ion pairing near these and the adjacent carbons, C-3 and C-3′. The electrostatic
potential calculated for the cation confirms the location of the preferred binding site.
In tr od u ction
drogen bond with the use of another acceptor, OPPh3,
failed. The crystallographic structural analysis showed
that the hydrogen on the nitrogen and a fluorine of the
counteranion are close, in the range of possible hydrogen
bonding. In other work,6 one of the fluorines of the
counteranion BF4 is found to hydrogen bond to a proton
of two different NH groups, which are themselves
hydrogen bonded to a hydride, making a hydrogen-
bonded four-membered ring. This interaction is seen in
the X-ray structure analysis as well as in the low-
temperature 1H NMR, but is not seen in the low-
temperature 19F NMR.
Even though ion pairing is recognized in coordination
and organometallic compounds and is known to influ-
ence reaction rate,1,2 it is often ignored in the majority
of discussions of organometallic structure and mecha-
nism. Metallocene polymerization is one area where ion
pairing both has a clearly important mechanistic role
and has been carefully studied.3 The structure of ion
pairs has sometimes been inferred from X-ray structural
work, where it is assumed that the solution structure,
the only structure relevant to reactivity in solution, is
the same as in the solid.
Macchioni, Ruegger, Venanzi, and their co-workers
have applied 1H NOESY and 19F{1H} HOESY NMR
spectroscopy to the investigation of predominant ion
pair solution structures of organometallic complexes by
the detection of interionic contacts.7,8 The results sug-
gest that these ion pairs have surprisingly well-defined
structures, often broadly similar to the solid-state
structure. For example, in one case, the anion is
surrounded in the solid state by six symmetry-related
cations, only one of which reflects the solution structure.9b
In such a case, even with the solid-state structure in
hand, it would not be obvious which of the ion pairs seen
in the solid would resemble the solution structure. One
An early finding of the possible influence of the
counteranion on the stability of compounds was made
by Bianchini et al.4 Depending on the counteranion
used, they obtained in the solid state either a dihydro-
gen cobalt complex with BPh4 or a dihydride cobalt
complex with PF6, whereas in solution only the dihy-
drogen isomer is seen for both counteranions.
Morris and co-workers detected a “bifurcated” hydro-
gen bonding involving a hydride, a proton from a
pendant neighboring amino group, and a fluorine of the
counteranion BF4, Ir-H‚‚‚H(N)‚‚‚F-B.5 Indeed, a previ-
ous attempt to disrupt the expected Ir-H‚‚‚H-N hy-
* Corresponding authors. E-mail: alceo@unipg.it, clot@lsd.univ-
montp2.fr, robert.crabtree@yale.edu.
(5) Park, S.; Lough, A. J .; Morris, R. H. Inorg. Chem. 1996, 35, 3001.
(6) Xu, W.; Lough, A. J .; Morris, R. H. Inorg. Chem. 1996, 35, 1549.
(7) Bellachioma, G.; Cardaci, G.; Macchioni, A.; Reichenbach, G.;
Terenzi, S. Organometallics 1996, 15, 4349.
† Universita` di Perugia.
‡ Universite´ de Montpellier II.
§ Yale University.
(1) Romeo, R.; Arena, G.; Scolaro, L. M.; Plutino, M. R. Inorg. Chim.
Acta 1995, 241, 81.
(2) Bellachioma, G.; Cardaci, G.; Macchioni, A.; Zuccaccia, C. J .
Organomet. Chem. 2000, 594, 119.
(3) Chen, Y. X.; Metz, M. V.; Li, L. T.; Stern, C. L.; Marks, T. J . J .
Am. Chem. Soc. 1998, 120, 6287.
(4) Bianchini, C.; Mealli, C.; Peruzzini, M.; Zanobini, F. J . Am.
Chem. Soc. 1992, 114, 5905.
(8) Macchioni, A.; Bellachioma, G.; Cardaci, G.; Gramlich, V.;
Ru¨egger, H.; Terenzi, S.; Venanzi, L. M. Organometallics 1997, 16,
2139.
(9) (a) Macchioni, A.; Bellachioma, G.; Cardaci, G.; Travaglia, M.;
Zuccaccia, C.; Milani, B.; Corso, G.; Zangrando, E.; Mestroni, G.;
Carfagna, C.; Formica, M. Organometallics 1999, 18, 3061. (b) Mac-
chioni, A.; Bellachioma, G.; Cardaci, G.; Cruciani, G.; Foresti, E.;
Sabatino, P.; Zuccaccia, C. Organometallics 1998, 17, 5549.
10.1021/om010015l CCC: $20.00 © 2001 American Chemical Society
Publication on Web 05/01/2001

2368 Organometallics, Vol. 20, No. 11, 2001
Macchioni et al.
trend evident from prior work7,8,9b,10 is that the anion
tends to prefer to remain in the vicinity of any N-
heterocyclic ligand, such as a pyridyl or pyrazolyl group.
It was not clear if this would continue to be true if the
complex had a cis-MH2 group present. Such a sterically
small group, also potentially able to hydrogen bond to
the anion, might attract the counterion because in this
way it could approach much more closely to the central
metal. Prior work had not used computational methods
to characterize the PE surface around the cation and
predict the ion pair structure.
In our work on Ir hydrides, we have seen anion effects
that made us consider ion pairing as an issue. One goal
of this paper was to find the solution ion pair structure
for [IrH2(bipy)(PRPh2)2]A (1,2, bipy ) 2,2′-bipyridyl; R
) Ph, A ) BF4, PF6, CF3SO3, and BPh4; R ) Me, A )
PF6) and compare it with the solid-state structures
(1b,c). As the smallest ligand, a hydride might be
thought to favor ion pairing in its vicinity so that the
metal‚‚‚anion distance would be minimized. Cationic
metal hydrides can be weakly acidic and can even
hydrogen bond, as in M-H‚‚‚base interactions,11 so the
ion pairing might no longer take place as in prior cases
near the bipy ligand. We find that both 1b and 1c have
dissimilar solid and solution structures, but in both
cases the anion prefers the vicinity of the bipyridyl
ligand to the vicinity of the hydrides, where it could
approach much closer to the central metal. Consistent
with experiment, theoretical work locates the positive
charge on the inter-ring carbons, C-2 and C-2′, of the
bipy ligand and not on the metal or bipy nitrogens as
previously assumed.
F igu r e 1. Section of the 1H-NOESY NMR spectrum of
complex 1d recorded at 400.13 MHz in methylene chloride-
d2 (298 K) showing the selective intramolecular interactions
of the hydrides with protons H-6 and H-o.
Ch a r t 1
Resu lts a n d Discu ssion
Syn th esis of Com p ou n d s 1a -d a n d 2. Pignolet et
al.12,13 and Crabtree et al.14 previously reported the
syntheses of complexes 1a -c. 1a was obtained by the
reaction of the bis-solvento complex [Ir(PPh3)2(acetone)2-
(H)2]BF4 and 2,2′-bipyridyl, and 1b by metathesis of 1a
with NH4PF6.12,13 Here, we report a modified procedure
for the preparation of complexes 1 and 2. A methylene
chloride solution of [Ir(cod)(PRPh2)2]X and 2,2′-bipyridyl
was treated with dihydrogen gas at 0 °C to give
cis,trans-[IrH2(bipy)(PRPh2)2]X (1, 2) in good yields.
Complexes 1 and 2 were fully characterized by elemen-
scalar and dipolar nuclear interactions in the 1H COSY,
1H, 13C HMQC, and 1H NOESY experiments, respec-
tively. The assignment of the protons of the bipy rings
was carried out starting from the observation of an NOE
contact between proton H-6 and hydride H-1′′ (Figure
1). The 1H NMR spectrum of compound 2 displays a
triplet resonance at -19.87 ppm (2J HP ) 16.8 Hz) for
the two equivalent hydrides H-1′′ and H-2′′, a signal at
1.88 ppm for the methyl protons, and a complex mul-
tiplet ranging from 7.08 to 8.46 ppm for the phenyl and
bipyridyl protons. All numbering refers to Chart 1.
NMR In ter ion ic In vestigation s. The relative anion-
cation position (interionic structure)10 of complexes 1
and 2 was investigated in methylene chloride-d2 by
dissolving about 20 mg of compound in 0.6 mL of
solvent. Under these conditions, the complexes are
mainly present in solution as intimate ion pairs,15 and
dipolar interactions between nuclei belonging to the two
1
tal analysis and IR, H, 13C, 31P, and 19F NMR spec-
troscopies. The spectroscopic data of 1a -c are consistent
with those in the literature.12-14 The data for the new
complexes 1d and 2, presented below, are closely similar
to those of 1a -c.
NMR In tr am olecu lar Ch ar acter ization . The struc-
ture of complexes 1 and 2, shown in Chart 1, was
1
investigated in methylene chloride-d2 by H, 13C, 31P,
and 19F NMR spectroscopies. The assignments of all the
1H and 13C resonances were performed following the
(10) Zuccaccia, C.; Bellachioma, G.; Cardaci, G.; Macchioni, A.
Organometallics 1999, 18, 1, and references therein.
(11) Desmurs, P.; Kavallieratos, K.; Yao, W.; Crabtree, R. H. New
J . Chem. 1999, 23, 1111.
(12) Alexander, B. D.; J ohnson, B. J .; J ohnson, S. M.; Casalnuevo,
A. L.; Pignolet, L. H. J . Am. Chem. Soc. 1986, 108, 4409.
(13) Alexander, B. D.; J ohnson, B. J .; J ohnson, S. M.; Boyle, P. D.;
Kann, N. C.; Pignolet, L. H. Inorg. Chem. 1987, 26, 3506.
(14) Guzei, I. A.; Rheingold, A. L.; Lee, D.-H.; Crabtree, R. H. Z.
Kristallogr.-New Cryst. Struct. 1998, 213, 585.
1
ionic moieties can be detected in the H NOESY or 19F-
{1H} HOESY spectra. As shown in Figure 2, the nuclei

Selective Ion Pairing in [Ir(bipy)H2(PRPh2)2]A
Organometallics, Vol. 20, No. 11, 2001 2369
Ta ble 1. Qu a n tifica tion of th e An ion -Ca tion
In ter a ction s
1a a
1ba
1ca
1d b
3
4
5
6
1.00
0.33
0.11
0.06
1.00
0.38
0.16
0.06
1.00
0.57
0.27
0.11
1.00
0.71
0.45
0.24
a
Values derived from the integration of NOE “contacts” nor-
b
malized with respect to protons H-3,3′. Values derived from the
shielding exerted by the π-electrons17a of BPh4 normalized with
-
respect to the ∆δ(1a -d ) value of protons H-3,3′. The ∆δ values
were raised to the 1/2 power in order to be quantitatively
comparable with the data from NOE measurements.
they can be useful to understand how the strength of
X-/aromatic proton interactions, relative to that of
X-/proton 3, is affected by the change of counteranion.
No contact is observed with the two hydrides H-1′′ and
H-2′′. This is true independent of the nature of the
counterion. The interactions are practically the same
in complex 2, with the addition of a weak interaction
between the phosphine Me protons and PF6-. The
observed anion-cation “contacts” afford a well-defined
structure for the ion pair: the anion is located close to
protons H-3,3′, as reported in Chart 1. The expected
localization of the counterion close to the hydride region,
which should allow a minimal metal anion distance, is
not observed. This suggests that the positive charge is
not localized on iridium, but is predominantly delocal-
ized onto the bipyridyl ligand, as already observed in
previous studies, where the nitrogens were considered
as the predominant positive centers.9 In complex 1d ,
protons close to the counteranion are shielded by the
π-electrons17a of the BPh4- aromatic rings, as expected
from the NOE-deduced interionic structure. For a
quantitative comparison with the data from NOE
measurements the ∆δH (ppm) data, normalized with
respect to protons H-3,3′, should be raised to the half
power to take into account that the shielding effect is a
first-order dipole-dipole interaction ( 1/r3),17a while the
NOE is a second-order relaxation effect ( 1/r6). The ∆δH
(ppm) of the bipy protons for 1a -d perfectly follows the
trend of NOE results (Table 1) with the degree of
interaction falling in the order 3 > 4 > 5 > 6. Further-
more, from Table 1 it can be concluded that the highest
specificity of anion-cation interactions is reached with
the smallest anion (BF4-). As an example, the intensity
of the interaction between the counteranion and the H-4
proton relative to that with H-3 increases with increas-
ing anion size: 0.33 (1a ) < 0.38 (1b) < 0.57 (1c) < 0.71
(1d ). The size of the counteranion plays a role in its
location around the cation, as well as the strength of
the interaction with the cation.
F igu r e 2. 19F{1H} HOESY NMR spectrum of complex 1a
recorded at 376.63 MHz in methylene chloride-d2 (298 K)
showing (a) the absence of the “expected” interionic NOE
between the hydrides and the anion and (b) the contacts
between BF4- and aromatic protons. In the expansion the
point (b) is better illustrated and the interaction intensity
order 3 > 4 > 5 > 6 is shown. The 1D-trace relative to the
11BF4 column is reported on the right of both total
-
spectrum and expansion.
of the counteranion of complexes 1 interact with protons
H-3,3′ (strong), H-4,4′ (medium), H-5,5′ (weak), H-m
(very weak), H-o (very weak), and H-6,6′ (very weak).16
The intensities of the interionic NOE interactions are
reported in Table 1, by arbitrarily fixing, in every
complex, the highest value to 1 (X- with proton 3). Even
though the data of Table 1 have little absolute meaning,
X-r ay Cr ystallogr aph ic Ch ar acter ization of Com -
p lexes 1b a n d 1c. The structures of complexes [IrH2-
(bipy)(PPh3)2]A 1b (A ) PF6) and 1c (A ) CF3SO3) were
previously reported (Figures 3 and 4).13,14 In both
compounds, the ligands around the central metal are
found in an essentially octahedral arrangement, with
the bipyridyl and the two hydride ligands in the plane
of the molecule. Distances are normal. The outer sphere
A- anions are located close to the H-6 and H-5 positions,
but far away from H-6′ and H-5′ positions of the
(15) Zuccaccia, C.; Bellachioma, G.; Cardaci, G.; Macchioni, A.
Organometallics 2000, 19, 4663. Bellachioma, G.; Cardaci, G.; Mac-
chioni, A.; Zuccaccia, C. Unpublished conductometric results relative
to octahedral Ru(II) complexes.
(16) From the 1D-traces reported in Figure 2 it could appear that
NOEs relative to the interionic interactions of o and m protons with
-
BF4 have the same intensity as that of protons H-4,4′. In reality, it
must be considered that for
a quantitative comparison the NOE
intensity has to be divided by the number of equivalent protons, i.e.,
12 and 2 for o/m and bipy protons, respectively. The intensity of the
interionic interaction of the o and m protons becomes similar to that
of protons H-4,4′.
(17) (a) Haigh, C. W.; Mallion, R. B. Prog. NMR Spectrosc. 1980,
13, 303. (b) Harris, R. K. Can. J . Chem. 1964, 42, 2275.

2370 Organometallics, Vol. 20, No. 11, 2001
Macchioni et al.
Ta ble 2. Selected Non bon d ed In ter ion ic Dista n ces
(Å) for Com p ou n d 1b a n d 1ca
1b
1c
H1′′-F6
H2′′-F6
3.068
5.002
H1′′-F2
H2′′-F2
H1′′-O2
H2′′-O2
H6-O2
H5-O2
H4-O2
H3-O2
H6-O3
H5-O3
H4-O3
H3-O3
H6′-O2
5.488
7.419
4.154
6.423
2.843
3.871
6.171
7.313
3.215
2.509
4.744
6.516
8.536
H6-F1
H5-F1
H4-F1
H3-F1
2.957
3.744
6.042
7.272
H6′-F1
8.709
a
Calculated from the crystal structures with H ligands in
calculated positions. Numbers refer to Chart 1.
bipyridyl ligand, and close to the hydride H-1′′, but far
away from the hydride H-2′′. Table 2 shows a selection
of the shortest nonbonded interionic distances, which
were obtained from the crystal structures of 1b and 1c
registered in the Cambridge Structural Database Sys-
tem. The expected localization of the counteranion close
to both hydrides is not observed; indeed the counteran-
ion is located close to one hydride and to one side of the
F igu r e 3. View of the solid-state structure of complex 1b
generated from CSDS. Crystallographic numbering applies.
hydridic character. Moreover the charge on iridium is
also negative, with a value of -0.13. This negative
charge is a combination of two effects. The phosphines
(PH3) are rather basic; thence some electron density is
transferred to the Ir. As a result, the global charge on
each PH3 ligand is +0.42. The other contributing effect
to the negative charge on Ir is π-donation from the
nitrogen atoms; indeed the NPA occupation of the 2pz
orbital (perpendicular to the bipy plane) is only 1.28,
indicating π-donation from nitrogen to iridium.
-
bipyridyl ligand. However, the counteranion PF6 is
closer to the metal center than CF3SO3-, which was
-
expected since CF3SO3 is larger than PF6-; thence
CF3SO3- cannot approach the metal center as easily due
to steric interactions with nearby bipyridyl and phos-
phine ligands. This anion-cation specificity was also
encountered in solution.
However, both nitrogen atoms are negatively charged
(-0.55) as a consequence of the bonding interactions
with neighboring positively charged carbon atoms C-2
(+0.21) and C-6 (+0.09). The other carbon atoms of the
bipy ring are negatively charged: -0.24 for C-5, -0.16
for C-4, and -0.22 for C-3. The negative charge on
nitrogen is mainly a consequence of σ-type interactions
causing a drift of electron density from the adjacent
atoms to nitrogen. The total NPA population of orbitals
2s, 2px, and 2py, from which the sp2 hybrids are
constructed, is 4.22, leading to a “σ-charge” of -1.22
compensated by a “π-charge” of +0.72 to give an overall
charge of ca. -0.5. Although σ-type density originates
from Ir, C-2, and C-6, the π-donation from nitrogen is
essentially directed toward iridium.
This charge analysis clearly shows that the positive
charge is located on the phosphine ligands and on the
carbon atoms C-2 and C-2′ that form the inter-ring C-C
bond of the bipy ligand. The use of PH3 as a model
phosphine is a limitation here, and the real phosphines,
PPh3, are certainly less positively charged. Even if the
P atoms bear some positive charge, they will be buried
in the bulk of the three phenyl rings and would not be
involved in any strong interaction with the counteran-
ion. However the presence of positive charge on carbons
C-2 and C-2′ is in very good agreement with the NMR
experiment and is clearly consistent with the location
of the counteranion.
Solid -Sta te vs Solu tion In ter ion ic Str u ctu r es of
Com p lexes 1a a n d 1c. The analysis of the spectro-
scopic data obtained in solution leads to the conclusion
that the counteranion is located close to protons H-3 and
H-4 of the bipyridyl ligand, far away from the two
hydrides, whereas in the solid state the counteranion
is found close to one of the hydrides (H-1′′) and close to
protons H-6 and H-5 of the bipyridyl ligand.
To elucidate this difference, we carried out hybrid
quantum mechanics (QM)/molecular mechanics (MM)
calculations on the complex Ir(H)2(bipy)(PPh3)2+. Den-
sity functional theory (B3PW91) calculations were car-
+
ried out on Ir(H)2(bipy)(PH3)2 as a QM model, and
phenyl substituents were treated at the MM level (UFF
force field) within the ONIOM methodology (see Com-
putational Details for further information).
The optimized geometry (ONIOM) of Ir(H)2(bipy)-
+
(PPh3)2 is shown in Figure 5, and some geometrical
parameters are given in Table 3 for comparison with
the published X-ray data for [Ir(H)2(bipy)(PPh3)2][CF3-
SO3]14 and [Ir(H)2(bipy)(PPh3)2][PF6].13 The ONIOM
geometry is in very good agreement with the experi-
mental data, and the disymmetry between the two Ir-N
bonds is particularly well reproduced. π-Stacking in-
teractions are present between the pyridine and phenyl
rings, and these stabilizing interactions are most likely
responsible for the distortion. The simulation of such a
subtle effect with ONIOM clearly illustrates the sensi-
tivity and the strength of this hybrid QM/MM method.
We performed a natural population analysis (NPA)
which shows that the positive charge is not localized
on the IrH2 part of the molecule. The NPA charges on
the hydrides are -0.015 and -0.026, indicating a weak
We also computed the electrostatic potential created
by the molecule, and a contour plot in the plane of the
bipy ligand is shown as Figure 6. As expected for a
positively charged molecule, the electrostatic potential
is positive, leading to global attraction for an anion.

Selective Ion Pairing in [Ir(bipy)H2(PRPh2)2]A
Organometallics, Vol. 20, No. 11, 2001 2371
F igu r e 4. View of the solid-state structure of complex 1c generated from CSDS. Crystallographic numbering applies.
Ta ble 3. Com p a r ison betw een Selected
Exp er im en ta l13,14 a n d Ca lcu la ted Bon d Dista n ces
+a
(Å) a n d An gles (d eg) for Ir (H)2(bip y)(P P h 3)2
ref 13
ref 14
QM/MM
Ir-N(1)
2.169
2.130
2.177
2.146
2.205
2.146
1.577
1.573
1.364
1.361
1.477
75.5
Ir-N(1′)
Ir-H(1′′)
Ir-H(2′′)
N(1)-C(2)
N(1′)-C(2′)
C(2)-C(2′)
N(1)-Ir-N(1′)
P-Ir-P
1.369
1.374
1.470
76.6
1.362
1.359
1.491
76.6
161.8
167.3
165.6
a
QM/MM refers to the geometry optimized within the ONIOM
methodology. The numbering of the atoms is the same as in Chart
1.
in agreement with the charge analysis, which gave
negative charges for IrH2.
The structure of the ion pair in solution as observed
by NMR spectroscopy is thus the result of the preferred
electrostatic interaction between the cation and the
anion, as illustrated by the orientation of the dipole
moment (Figure 5) and the topology of the electrostatic
potential (Figure 6). The solid-state structure might
then be a compromise between this preferred geometry
for the ion pair and a more compact and more stable
arrangement in the solid state. The liquid state being
less constrained, the preferred electrostatic interaction
is the leading contributor to the ion pair formation.
F igu r e 5. (a) Optimized geometry (ONIOM) for Ir(H)2-
(bipy)(PPh3)2+. (b) Schematic representation of the QM part
in the ONIOM calculations (PH3 groups have been omitted
for clarity) showing the orientation of the dipole moment
of the molecule.
Con clu sion s
Two-dimensional homo- and heteronuclear NMR spec-
troscopies detect ion pairing for 1 and 2 in solution and
identify the preferred structure of the ion pair. Instead
of binding near the MH2 group, which would give the
closest M‚‚‚anion distance and allow hydrogen bonding,
the anion is located near the dipy ligand. The ion pairing
is enhanced for smaller anions where the counteranion
can approach closer to the cation. Since the solution and
solid-state ion pair structures are different, crystal-
lographic characterization has limited relevance to the
However, the topology of the potential is in favor of a
preferred interaction in the vicinity of C-2 and C-2′.
There is a smooth attractive valley pointing toward the
midpoint of the C-2-C-2′ bond, in perfect agreement
with the NMR experiment. Moreover the electrostatic
potential clearly shows that the IrH2 region is the less
attractive region despite its lower steric bulk. This is

2372 Organometallics, Vol. 20, No. 11, 2001
Macchioni et al.
3
3
7.38 (d, J o,m ) 7.2, o), 7.35 (t, J p,m ) 8.2, p), 7.28 (t, (3J m,p
+
3J m,o)/2 ) 7.5, m), 6.79 (ddd, J 3,4 ) 7.6, J 3,2 ) 5.5, J 3,5 ) 1.3,
3
3
4
3), - 19.45 (t, 2J H,P ) 16.6, H). 13C{1H} NMR (CD2Cl2, 298 K):
2
δC (ppm) 156.0 (s, 6), 155.3 (s, 2), 137.5 (s, 4), 133.5 (t, J o,P
)
6.1, o), 131.7 (t, 1J ipso,P ) 27.0, ipso), 130.7 (s, p), 128.7 (t, 3J m,P
) 5.0, m), 127.0 (s, 3), 123.8 (s, 5). 31P{1H} NMR (CD2Cl2, 298
K): δP (ppm) 21.4 (s, PPh3). 19F NMR (CD2Cl2, 298 K): δF (ppm)
- 153.50 (br,10BF4-), - 153.55 (q,1J F,B ) 1.0, 11BF4-). IR (thin
film, cm-1): ν(Ir-H) 2155.5, ν(BF4) 1073.6. Anal. Calcd for
C
46H40IrN2P2BF4‚0.25CH2Cl2: C, 56.51; H, 4.15; N, 2.85.
Found: C, 56.48; H, 4.13; N, 2.80. Yield: 83.6%.
Ch a r a cter iza tion of cis,tr a n s-[Ir H2(bip y)(P P h 3)2]P F 6
1
3
(1b). H NMR (CD2Cl2, 298 K): δH (ppm) 8.17 (d, J 2,3 ) 5.3,
3
3
3
2), 7.91 (d, J 5,4 ) 8.2, 5), 7.76 (ddd, J 4,3 ) 7.4, J 4,5 ) 8.1,
4J 4,2 ) 1.4, 4), 7.35 (m, o and p), 7.27 (t, (3J m,p + 3J m,o )/2 ) 7.3,
m), 6.80 (ddd, 3J 3,4 ) 7.6, 3J 3,2 ) 5.5, 4J 3,5 ) 1.1, 3), - 19.44 (t,
2J H,P ) 16.8, H). 13C{1H} NMR (CD2Cl2, 298 K): δC (ppm) 156.0
(s, 6), 155.4 (s, 2), 137.4 (s, 4), 133.5 (t, 2J o,P ) 6.0, o), 131.7 (t,
1J ipso,P ) 26.9, ipso), 130.7 (s, p), 128.7 (t, 3J m,P ) 5.0, m), 127.1
(s, 3), 123.6 (s, 5). 31P{1H} NMR (CD2Cl2, 298 K): δP (ppm)
1
21.4 (s, PPh3), - 143.16 (sept, J P,F ) 711.0, PF6-). 19F NMR
(CD2Cl2, 298 K): δF (ppm) - 73.71 (d,1J F,P ) 711.0, PF6-). IR
(thin film, cm-1): ν(Ir-H) 2150.4, ν(PF6) 836.9. Anal. Calcd
for C46H40IrN2P3F6‚0.5CH2Cl2: C, 52.81; H, 3.91; N, 2.65.
Found: C, 53.00; H, 3.78; N, 2.57. Yield: 84.1.
F igu r e 6. Contour plot of the electrostatic potential in the
plane of the bipy ligand. Contour lines are from 0.08 to 0.3
au with a 0.02 increment between two consecutive con-
tours.
Ch ar acter ization of cis,tr a n s-[Ir H2(bipy)(P P h 3)2]CF3SO3
1
3
(1c). H NMR (CD2Cl2, 298 K): δH (ppm) 8.16 (d, J 2,3 ) 5.4,
3
3
3
2), 8.04 (d, J 5,4 ) 8.2, 5), 7.79 (ddd, J 4,3 ) 7.6, J 4,5 ) 8.2,
4J 4,2 ) 1.5, 4), 7.38 (d, J o,m ) 7.2, o), 7.35 (m, o and p), 6.79
3
3
3
4
2
(ddd, J 3,4 ) 7.6, J 3,2 ) 5.5, J 3,5 ) 1.2, 3), -19.45 (t, J H,P
)
16.8, H). 13C{1H} NMR (CD2Cl2, 298 K): δC (ppm) 156.1 (s, 6),
155.3 (s, 2), 137.5 (s, 4), 133.5 (t, 2J o,P ) 6.1, o), 131.7 (t, 1J ipso,P
) 27.0, ipso), 130.7 (s, p), 128.7 (t, 3J m,P ) 4.9, m), 127.0 (s, 3),
123.9 (s, 5). 31P{1H} NMR (CD2Cl2, 298 K): δP (ppm) 21.4 (s,
PPh3). 19F NMR (CD2Cl2, 298 K): δF (ppm) - 79.27 (s,
CF3SO3-). IR (thin film, cm-1): ν(Ir-H) 2194.2. Anal. Calcd
for C46H40IrN2P2F2SO3: C, 55.13; H, 3.94; N, 2.73. Found: C,
54.90; H, 3.84; N, 2.64. Yield: 82.2%.
solution case. Differences in ion pairing could affect
structure and reactivity. This is of potential importance
since reactions are usually carried out in solution in
organic solvents, many of which favor ion pairing.
Finally, we show how theory can satisfactorily explain
the ion pair structure found in solution as a result of
postive charge buildup at the ring junction (C-2 and
C-2′) of the bipy ligand, not the N, as previously
assumed; in contrast, all the atoms of the H2IrN2
fragment bear a negative charge.
Ch a r a cter iza tion of cis,tr a n s-[Ir H2(bip y)(P P h 3)2]BP h 4
1
3
(1d ). H NMR (CD2Cl2, 298 K): δH (ppm) 8.13 (d, J 2,3 ) 5.3,
2), 7.55 (AB system, 3J 5,4 ) 8.4, 4 and 5), 7.37 (m, o, oX and p),
7.27 (m, m), 7.05 (t,3J m,o ) J m,P ) 7.3, mX), 6.90 (tt, J p,m
)
3
3
4
2
7.2, J p,o ) 1.2, pX), 6.70 (m, 3), - 19.46 (t, J H,P ) 16.8, H).
13C{1H} NMR (CD2Cl2, 298 K): δC (ppm) 164.5 (q, 1J C,B ) 49.2,
ipsoX), 155.7 (s, 6), 155.3 (s, 2), 137.3 (s, 4), 136.3 (s, oX), 133.5
(t, J o,P ) 6.1, o), 131.7 (t, J ipso,P ) 26.9, ipso), 130.7 (s, p),
128.7 (t, 3J m,P ) 5.0, m), 127.0 (s, 3), 126.0 (b, mX), 123.4 (s, 5),
122.1 (s, pX). 31P{1H} NMR (CD2Cl2, 298 K): δP (ppm) 21.4 (s,
PPh3). IR (thin film, cm-1): ν(Ir-H) 2171.1. Anal. Calcd for
Exp er im en ta l Section
All operations were carried out under argon atmosphere
using standard Schlenk techniques. Solvents were dried over
calcium hydride (CH2Cl2) or sodium/benzophenone (Et2O).
Hexane was degassed prior use. IR spectra were recorded on
a Midac M1200 FT-IR spectrometer. Microanalyses were
carried out by Robertson Microlit Laboratories. One- and two-
dimensional 1H, 13C, 19F, and 31P NMR spectra were measured
on a Bruker DRX 400 NMR spectrometer. Referencing is
relative to external TMS (1H and 13C), CFCl3 (19F), and 85%
H3PO4 (31P). NMR samples were prepared dissolving about 20
mg of compound in 0.6 mL of deuterated solvent. Two-
dimensional 1H NOESY and 19F{1H} HOESY spectra were
recorded with a mixing time of 500-800 ms.
cis,tr a n s-[Ir H2(bip y)(P RP h 2)2]X (1, 2). Typ ica l P r oce-
d u r e. F or 2: [Ir(cod)(PMePh2)2]PF6 (1 equiv, 286.3 mg, 0.338
mmol) and 2,2′-bipyridyl (1 equiv, 53.0 mg, 0.338 mmol) were
dissolved in degassed CH2Cl2 (10 mL). The resulting red
solution was cooled to 0 °C. Dihydrogen gas was bubbled for
20 min, with constant stirring. To the orange solution, diethyl
ether (100 mL) was added dropwise to precipitate 250.9 mg of
beige solid (82.9%), which was obtained by filtration in vacuo,
and dried in vacuo. Recrystallization is performed with CH2-
Cl2/Et2O.
2
1
C
70H60IrN2P2B‚1.5CH2Cl2: C, 64.98; H, 4.81; N, 2.12. Found:
C, 65.26; H, 4.61; N, 1.85. Yield: 68.1%.
Ch a r a cter iza tion of cis,tr a n s-[Ir H2(bip y)(P P h 2Me)2]-
1
3
P F 6 (2). H NMR (CD2Cl2, 298 K): δH (ppm) 8.46 (dd, J 2,3
)
5.4, 5J 2,5 ) 0.7, 2), 7.93 (d, 3J 5,4 ) 8.0, 5), 7.87 (ddd, 3J 4,3 ) 7.5,
3J 4,5 ) 8.2, J 4,2 ) 1.5, 4), 7.33 (m, p), 7.25 (m, o and m), 7.08
4
(ddd, J 3,4 ) 7.1, J 3,2 ) 5.5, J 3,5 ) 1.4, 3), 1.88 (t Harris,17b
3
3
4
| J H,P + 4J H,P| ) 6.4, PMePh2), -19.87 (t, 2J H,P ) 16.8, H). 13C-
2
{1H} NMR (CD2Cl2, 298 K): δC (ppm) 155.9 (s, 6), 155.4 (s, 2),
1
2
137.5 (s, 4), 133.4 (t, J ipso,P ) 26.7, ipso), 132.0 (t, J o,P ) 6.0,
3
o), 130.5 (s, p), 128.7 (t, J m,P ) 5.0, m), 127.5 (s, 3), 123.8 (s,
5), 17.0 (t, | J C,P + 3J C,P| ) 37.6, PMePh2). 31P{1H} NMR (CD2-
1
Cl2, 298 K): δP (ppm) - 0.27 (s, PMePh2), - 143.14 (sept, 1J P,F
) 711.0, PF6-). 19F NMR (CD2Cl2, 298 K): δF (ppm) - 73.50
(d,1J F,P ) 711.0, PF6-). IR (thin film, cm-1): ν(Ir-H) 2176.1,
ν(PF6) 836.2. Anal. Calcd for C36H36IrN2P3F6: C, 44.62; H, 3.84;
N, 2.74. Found: C, 44.58; H, 4.14; N, 2.75. Yield: 82.9%.
Str u ctu r a l Deta ils. The structures of complexes 1b and
1c have been fully reported.13,14 The nonbonded interionic
distances were obtained from the crystal structures of 1b and
1c, registered in the Cambridge Structural Database System.
The hydrogen atom positions of the bipyridyl ligand and the
Ch a r a cter iza tion of cis,tr a n s-[Ir H2(bip y)(P P h 3)2]BF 4
3
(1a ). 1H NMR (CD2Cl2, 298 K): δH (ppm) 8.16 (ddd, J 2,3
)
5.5, 4J 2,4 ) 1.6, 5J 2,5 ) 0.8, 2), 8.00 (ddd, 3J 5,4 ) 8.2, 4J 5,3 ) 1.1,
5J 5,2 ) 1.0, 5), 7.78 (ddd, J 4,3 ) 7.6, J 4,5 ) 8.2, J 4,2 ) 1.6, 4),
3
3
4

Selective Ion Pairing in [Ir(bipy)H2(PRPh2)2]A
Organometallics, Vol. 20, No. 11, 2001 2373
hydride ligand positions were calculated. Use of TEXSAN
software allowed the generation of the views of the solid-state
structure of the complexes and counteranions.
were also treated with Stuttgart’s RECPs and the associated
basis set,22 augmented by a polarization d function (R ) 0.387).
A 6-31G(d,p) basis set was used for the hydrides and for both
nitrogen atoms. The remaining atoms were treated by a 6-31G
basis set. The molecular mechanics calculations were per-
formed with the UFF force field.23
Com p u ta tion a l Deta ils. All calculations were performed
with the Gaussian 98 set of programs18 with the ONIOM
method.19 The complex Ir(H)2(bipy)(PPh3)2+ was optimized at
the ONIOM(B3PW91/UFF) level, where the QM part Ir(H)2-
(bipy)(PH3)2+ was optimized within the framework of density
functional theory at the B3PW91 level.20 The iridium atom was
represented by the relativistic effective core potential (RECP)
from the Stuttgart group (17 valence electrons) and its
associated (8s7p5d)/[6s5p3d] basis set.21 The phosphorus atoms
Natural population analysis (NPA)24 and calculation of the
electrostatic potential were performed on the QM part as
obtained from the ONIOM calculation, namely, model complex
+
Ir(H)2(bipy)(PH3)2 with a frozen geometry as deduced from
the ONIOM calculation. For these calculations Ir was treated
as before and all the remaining atoms were described with
the cc-pVDZ basis set of Dunning.25
(18) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.;
Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .; Barone, V.;
Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma,
K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J . B.;
Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, P. I.; Gomperts, G.; Martin, R. L.; Fox, D. J .;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez,
C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen, W.; Wong,
M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.; Pople J . A.
Gaussian 98; Gaussian, Inc.: Pittsburgh, PA, 1998.
(19) Svensson, M.; Humbel, S.; Froese, R. D. J .; Matsubara, T.;
Sieber, S.; Morokuma, K. J . Phys. Chem. 1996, 100, 19357.
(20) (a) Becke, A. D. J . Chem. Phys. 1993, 98, 5648. (b) Perdew, J .
P.; Wang, Y. Phys. Rev. B 1992, 82, 284.
(21) Andrae, D.; Ha¨ussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
Ack n ow led gm en t. This work was supported by
grants from the Ministero dell’Universita` e della Ricerca
Scientifica e Tecnologica (MURST, Rome, Italy), Pro-
gramma di Rilevante Interesse Nazionale, Cofinanzia-
mento 2000-1 (A.M., C.Z.), the CNRS and the University
of Montpellier II (E.C.), and the NSF (R.H.C., K.G.).
OM010015L
(22) Bergner, A.; Dolg, M.; Ku¨chle, W.; Stoll, H.; Preuss, H. Mol.
Phys. 1990, 30, 1431.
(23) Rappe, A. K.; Casewitt, C. J .; Colwell, K. S.; Goddard, W. A.;
Skiff, W. M. J . Am. Chem. Soc. 1992, 114, 10024.
(24) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J . Chem. Phys. 1985,
83, 735.
(25) Dunning, T. H., J r. J . Chem. Phys. 1989, 90, 1007.